Skip to main content

Full text of "Greenwich Time"

See other formats



and the discovery 
of the longitude 


Derek Howse 

This is the story of Greenwich Time, from 1676 
when it began to be used by two astronomers in 
the newly-founded Royal Observatory, to the 
present day when it has become the basis of the 
system of time kept the world over. It is also the • 
story of the finding of longitude at sea, for 
which specific purpose Charles II founded the 
Royal Observatory in his pa rk at G reen wich 
three hundred years ago. Greenwich became 
a household name when, in 1884, its Meridian - 
not that of Paris, or Washington, or the Great 
Pyramid - was chosen to be the world's prime 
meridian for longitude and time: Longitude Zero. 

Theauthor describes in simple terms the 
developments in astronomy, navigation, and 
timekeeping which matched increasing speeds 
of travel, and how Greenwich Time has been 
found, kept, and distributed. He also tells some 
of the many entertaining stories which 
accompanied this progress: how Columbus's 
prediction of an eclipse of the sun persuaded 
the natives to give him food at a crucial time; 
howa king's mistress had a hand in the 
founding of a national observatory; howa 
shipwreck(anda prize ofa quarter ofa million 
pounds in today's money) had such an influence 
on precision timekeeping; howdifficult it is to 
run a railroad if the times kept at both ends of 
the line are different; how the atomic clock is 
a more accurate timekeeper than the Earth itself; 
and howCo-ordinated Universal Time(the 
name given to the time-scale which is the basis 
of all the world's time signals) is still, whatever it 
may be called, firmly based on G reenwich 
Time, and is never more than a second different 
from it. 

Derek Howse is Head of Navigation and 
Astronomy at the National Maritime Museum, 
G reenwich, England. As such, he is responsible 
for the buildings of the Old Royal Observatory 
which now forms part of the museum (the 
astronomers having gone to Sussex), telling the 
story of astronomyand time. 

£7.95 net 
in UK 

Greenwich time 

and the discovery of the longitude 

Greenwich time 

and the discovery of the longitude 

Derek Howse 

Head of Navigation and Astronomy 

National Maritime Museum 

Oxford Nine York Toronto Melbourne 

Oxford University Press, Walton Street, Oxford 0x2 6dp 






© Derek Howse 1980 

All rights reserved. No part of this publication maybe 
reproduced, stored in a retrieval system, or transmitted, 
in any form or by any means, electronic, mechanical, 
photocopying, recording, or otherwise, without the prior 
permission of Oxford University Press 

British Library Cataloguing in Publication Data 
Howse, Derek 

Greenwich time and the discover,' of the longitude. 

" . Time measurements - History 

2. Greenwich, Eng. Royal Observatory • 

3- Longitude Measurement- History 
I. Title . • — 



BOt>K StVtfat 

Printed in Great Britain by Lowe and Brydone Printers Limited 
Thelford, Norfolk 


by Professor F. Graham Smith, FRS, Director, Royal 
Greenwich Observatory 

The first three hundred years of the Royal Greenwich 
Observatory were concerned primarily with the determin- 
ation of time and longitude for the purposes of navigation 
and surveying. Throughout this period there were, how- 
ever, other observatories measuring star positions and 
using the Earth's rotation as a clock. How did it come about 
that Greenwich achieved the distinction of defining longi- 
tude zero and giving its name to Greenwich Mean Time? 
The story is partly due to successive Astronomers Royal, 
especially Maskelyne and Airy, but the story also has its 
chance elements, and its amusing side, all of which make it 
the more worth telling. It is told in this book far better than 
ever before. 

The recent move of the Observatory from Greenwich to 
Herstmonceux coincides with a shift of interest towards 
astrophysics, and a shift of purpose towards the support of 
university research groups. The Observatory continues, 
however, to enjoy a unique reputation for a combination of 
basic research and international collaboration which is 
based on the history of time and longitude, and it is proud to 
continue as the source of •tlie-'btiadcast time signals of 
Greenwich Mean Time. 


Foreword by Professor F. Graham Smith, FRS, 
Director of the Royal Greenwich Observatory 

Author's preface 

1 Seeking the longitude: 300 BC-AD 1675 
Longitude in ancient times 

The age of discovery 
Longitude by lunar distance 
The chronometer method invented 
The longitude prizes 
Seventeenth-century developments 

2 Greenwich time for astronomers: 1675 - 1720 
Setting the scene 

The foundation of Greenwich Observatory 
The building of Greenwich Observatory 
rhe Great Clocks 
Publication of results 

3 Greenwich time for navigators: 1700 - 1840 
The Longitude Act 

French prizes 

The invention of the sextant 

The Nautical Almanac 

The marine chronometer in Britain 

The marine chronometer in France 

The foundation of the French Bureau des Longitudes 


4 Greenwich time for Great Britain: 1825 - 1880 
1 ime for civil purposes 

Time distribution by hand 

Post Office time and railway time 






2 3 






The start of the Greenwich time service 

Procedure at the Royal Observatory 

Nationwide distribution of time signals 

The ETC's Strand time-ball 

The Deal time-ball 

Early developments in the time-distribution system 

Time signals to private subscribers 

Railway time P. local time 

Legal time 

5 A prime meridian 1790 - 1884 
The longitude of observatories 
The Atlantic cable 
European railways 

US railroads 
The prime meridian 
International discussions 
The International Meridian Conference, 

6 Greenwich time for the world: 1884 - 1939 
Standard time 

Unification of civil and astronomical days 

The International Date Line 

Radio time signals 

The Bureau International de I'Heure 

Daylight Saving Time 

Standard time at sea 

Greenwich time in the home 

The speaking clock 

The meridian moved 

7 A clock more accurate than the Earth 

Quartz crystal clocks 
The non-uniform Earth 
Ephemeris time 
The atomic clock 
Co-ordination of time signals 
The leap second 













I Finding the longitude 

Longitude by lunar eclipse 194 

Longitude by lunar distance 194 

Longitude by eclipses of Jupiter's satellites 197 

Longitude by chronometer 198 

II Time-finding by astronomy 

The method 199 

The instruments 200 

III Mechanical and electrical clocks by Roger Stevenson 

The basis of a clock 205 

Tlie verge escapemen t 206 

The pendulum clock 207 

The recoil escapement 208 

Tlamsteed's great clocks 209 

The dead-beat escapement and mercurial pendulum 209 

John Harrison 's gridiron pendulum 212 

Further improvements 212 

Electrical timekeeping 213 

Dent's regulator 215 

The Riefler escapement 215 

The free-pendulum clock 216 

IV Modern precision clocks by John Pilkington 220 

V Time-balls in operation, 1861 

List of time-balls in actual operation, 1861 May 227 
List of places at which time-balls have been 

projected 228 

VI International Meridian Conference, Washington, 

I ist of delegates 229 

Bibliography 232 

Notes 235 

Index 247 





Abbreviations: NMM, National Maritime Museum; RGO, Royal 
Greenwich Observatory; BL, British Library; BM, British Museum; 
RAS, Roval Astronomical Society 

1 Regiomontanus's almanac, 1474. RAS 

2 The cross-staff in use. BL 

3 The earliest illustration of the lunar-distance method 
of finding longitude. Bodleian Library, Oxford 

4 The chronometer method of finding longitude - an 
English description of 1559. NMM 

5 France re-surveyed. BL 

6 Louise de Keroualle. Courtaidd Institute of Art 

7 Sir Jonas Moore. NMM 

8 John Flamsteed, first Astronomer Royal, with his 
assistant Thomas Weston. NMM 

9 Greenwich Observatory from Croom's Hill, about 
1680. NMM 

10 The equation of time 

11 The Great Room, Royal Observatory, Greenwich, 
about 1676. NMM 

12 John Flamsteed, 1684. RGO 

13 Flamsteed's star catalogue. 1 NMM 

14 The wreck of Sir Clowdisley Shovel's fleet, 1707. 

15 William Whiston. B. and E. Nicholts 

16 The longitude lunatic (detail from Hogarth's The 
Rake's Progress). BM 

17 Lunar-distance table, 1772. NMM 

18 Taking a lunar distance. NMM 

19 Harrison's first marine timekeeper of 1735 (Hi), his 
prizewinning watch H4 (1759), Kendall's copy (Ki, 
1769), and Kendall's second (K2, 1771). NMM 











20 Kendall's Ki. NMM 70 

21 The Greenwich time-ball of 1833. NMM 76 

22 Le Roy's 'montre marine', 1766. Conservatoire 

National des Arts et Metiers, Paris 76 

23 Nineteenth-century train guards' watches. Crown 
copyright. National Raihvai/ Museum, York 85 

24 Miss Ruth Belville. RGO 85 

25 Greenwich Observatory about 1870. NMM 93 

26 Shepherd's mean solar standard clock and 

apparatus for dropping the time-ball, 1885. NMM 93 

27 The time-ball in Comhill, London, i860. RGO 95 

28 Single-needle telegraph instrument, used for 

sending and receiving time signals. NMM 95 

29 The Chronopher Room at the Central Telegraph 
Office, General Post Office, St.-Martins-le-Grand, 

1874. NMM 95 

30 The time-ball in the Strand, London, 1852. MvlM 98 
31, 32 Small time-ball, about 1855, and galvanometer, 

about 1900. NMM 103 

33 A hatter's time signal at George Carter & Sons Ltd., 
Old Kent Road, London. S.E. London and Kentish 
Mercun/ 103 

34 Sir George B. Airy, KCB, FRS, from Punch, 1883. 
NMM 103 

35 Towns in Great Britain still keeping local time in 
February 1852. RGO 110 

36 Tom Tower and the Great Clock at Christ Church, 
Oxford, about i860. Bodleian Libran/, Oxford 1 1 1 

37 The Great Eastern off Valentia Island before laying 

the Atlantic cable, 1865. NMM 1 19 

38 Splicing the Atlantic cable on board the Great 
Eastern, 1865. NMM 119 

39 Charles F. Dowd, of Saratoga Springs, NY, 1883. BL 122 

40 US time zones, in 1883 and today 125 

41 Multiple longitude scales on the charts in the official 
French atlases, Le Neptune francois. NMM 130 

42 SirSandford Fleming. Canadian Pacific Corporate 

Archhxs 133 

43 Airy's transit circle at Greenwich, 1880. NMM 143 
Paris Observatory. Observatoire de Paris - /. Cotiml 143 

45 The civil, astronomical, and nautical days 149 


46 Time zone chart, 1979. HMSO ami Hydrographer of the 
Navy 158-9 

47 Wireless time signals at sea, 1910. Marconi Inter- 
national Marine Co. Ltd. 168 

48 Wireless time signals in Britain in the home, 1923. 
Marconi Co. Ltd. 168 

49 The Earth as a clock, showing changes in the Earth's 
rotation rate over the last 300 years. RGO 176 

50 Polar wander, 1971-7. RGO 177 

51 The increasing accuracy of precision clocks. RGO 181 

52 Time signals and the leap second. RGO 185 

53 Time Service Control Room, Royal Greenwich 
Observatory, Hcrstmonceux. RGO 188 

54 The Greenwich time-ball today. .VMM 188 
55, 56 The Greenwich meridian today. NMM 189 

57 Longitude and time 193 

58 Lunar-distance changes, relative to the Sun or to a 
zodiacal star 195 

59 Lunar-distance observations 196 

60 A transit instrument 201 

61 Photographic zenith tube - general arrangement. 

RAS 202 

62 The Greenwich photographic zenith tube, 1974. 

RGO 203 

63 The Greenwich PZT console. NMM 203 

64 A Danjon impersonal prismatic astrolabe at Paris 
Observatory. Observatoirc de Paris 203 

65 Clock elements 205 

66 Verge escapement 206 

67 Verge escapement controlled by a pendulum 207 

68 Recoil escapement 208 

69 The movement of a pendulum clock of 1768, by John 
Shelton. NMM 210 

70 Shortt free-pendulum clock No. 16, master and 
slaves, at Greenwich about 1930. NMM 210 

71 Dead-beat escapement 211 

72 Gridiron pendulum 213 

73 Shepherd's gravity escapement 214 

74 Free-pendulum slave movement 217 

75 Connections between free-pendulum and slave 
clocks 218 



76 Quartz crystal clock (diagram) 

77 Quartz crystal clock. RGO 

78 Caesium beam frequency standard, at I lerstmonceux, 
1974. RGO 

79 Caesium beam tube (diagram) 



Author's preface 

I. That it is the opinion of this Congress that it is desirable to 
adopt a single prime meridian for all nations, in place of the 
multiplicity of initial meridians which now exist. 

II. That the Conference proposes ... the adoption of the 
meridian passing through the centre of the transit instrument at the 
Observatory of Greenwich as the initial meridian for longitude. 

V. That this universal day is to be a mean solar day; is to begin 
for all the world at the moment of mean midnight of the initial 
meridian, coinciding with the beginning of the civil day and date of 
that meridian ..." 

Today, the name Greenwich is known and used by men and 
women of every race and creed all over the world. This is the 
result of the resolutions quoted above, resolutions taken on 
22 October 1884 at the end of the International Meridian 
Conference which met in Washington, DC, 'for the purpose 
of discussing, and, if possible, fixing upon a meridian 
proper to be employed as a common zero of longitude and 
standard of time-reckoning throughout the whole world'. 2 
This book tells the story of Greenwich time. It tells how, 
three hundred years ago, it was used only by those who 
lived and worked in Greenwich, perhaps a few hundred 
people, the most important from our point of view being 
two astronomers in the newly founded Royal Observatory 
in Greenwich Park; how, two hundred years ago, it began 
to be employed by seamen of all nations who used the 
newly published British Nautical Almanac, an annual publi- 
cation whose data were based on the Greenwich meridian; 
how, one hundred years ago, improved communications 
made it desirable to co-ordinate the time kept world-wide, 
resulting in the system of standard times used today, based 
on the Greenwich Meridian; how Greenwich Mean Time 


has today become Universal Time, the basis of the time 
employed for both domestic and scientific purposes all over 
the world - and why it was Greenwich time (and not, say, 
Paris time) that was chosen as the world's Universal Time. 

The need for such Universal Time, a time-scale of world- 
wide application and available to the world at large, has, 
however, arisen only quite recently. Since the earliest times 
Man has regulated his daily activities by the Sun - by its 
rising, its culmination (noon or midday), and its setting. 
Indeed, in many early civilizations, the day was divided 
into twelve 'hours of the day' (sunrise to sunset) and twelve 
'hours of the night' (sunset to sunrise), which were of differ- 
ent length. As summer days are longer than winter days, 
not only were the day-time hours of different length from 
the night-time hours (except at the equinoxes in March and 
September), but they also varied according to the seasons. 
Travellers discovered there were also variations according 
to how far one was from the equator. 

This system of 'unequal hours', based on local sunrise 
and sunset, with all its apparent complications, worked 
well enough for most domestic and business purposes 
except in the more northerly places where the discrepancy 
between night and day might be considerable. Indeed, it 
was still in use by some people in Italy until the fifteenth 
century. Greek astronomers, however, had divided their 
day into twenty-four hours of equal length and their 
successors of all nations continued this tradition. 

Whatever system of hours was used, however, the time 
kept by the ordinary person was local time, as shown by a 
sundial at the place where he was. To that ordinary person it 
was of no consequence that a place to the east or west of him 
actually kept different time by the Sun; that (because of the 
rotation of the Earth) when it was noon in London, it was 
11.44 in Plymouth but 12.05 ' n Norwich; or, put another 
way, that noon (when the Sun was due south) occurred in 
London 16 minutes before Plymouth but 5 minutes after 
Norwich. When Man's fastest speed of travel depended 
upon the horse, what did these differences matter? In any 
case, only in the last few hundred years has the keeping of 
time to any sort of accuracy been of any consequence in 
daily life, and then generally only in towns. 



author's preface 

Although the ordinary person might have no need for 
accurate time, there were some -geographers, map-makers, 
astronomers, travellers, and navigators - who did have 
such a need, not to find time for its own sake, but because, 
through time, they might be able to find, for example, the 
difference of longitude between places. How to do it was 
simple - in theory. Say you wanted to find the difference of 
longitude between London and Plymouth: if you measured 
the exact local time in London - say noon - and had some 
way of discovering the exact local time at the ven/ same 
moment in Plymouth - say 11.44 - 'hen the difference - 16 
minutes - gave you the difference in longitude between the 
meridians of London and Plymouth, 16 minutes of time 
being equivalent to four degrees of arc. And, because 
Plymouth noon occurs after London noon, Plymouth must 
be west of London, by 4°. (The problems of finding longi- 
tude are dealt with more fully in Appendix I.) 

But the gap between theory and practice was a large 
one. How could you find simultaneously the local times of 
an event as recorded at two widely separated places? Of 
course, there might be other methods of finding differences 
of longitudes - actual measurement of distances on the 
ground, or measuring the Earth's magnetism, for example - 
but the astronomical method, by finding the difference of 
times, seemed always to offer the best chance of success. 
The basic concept was known to the Greek astronomer 
I lipparchos by C. 180 bc but the accurate determination of 
longitude had to wait until the 1650s on land, and the 1770s 
at sea. 

The early part of this story of Greenwich time was thus 
entirely bound up with efforts to find a practical method of 
finding longitude, particularly at sea where it was made all 
the more difficult by the motion of the ship and the length of 
the voyages. This was no abstract concept: it was a highly 
practical necessity, as will be made clear. 


Much of the research for this book was done when pre- 
paring the displavs in the recently opened Time Galleries in 
the Old Royal Observatory. 
My principal thanks are due to my colleagues at the 

author's preface 

National Maritime Museum, in particular to Mr Basil 
Greenhill and Mr David Waters, respectively Director and 
Deputy Director, for their encouragement and for placing so 
many of the museum's facilities at my disposal; to Mrs 
Susan Gaston and Dr J. A. Bennett, Mr Alan Stimson, 
Mr Richard Good and Mr Roger Stevenson, and other 
members of the Department of Navigation and Astronomy 
past and present; and to Mr Brian Tremaine and his staff for 
producing so many of the photographs. I owe a particular 
debt to Mr Stevenson and his wife Mary who respectively 
wrote Appendix III and transformed my scribbles into 
impeccable typescript. 

The book would not have been possible without the 
readiest co-operation also from the staff of the Royal 
Greenwich Observatory at Herstmonceux, under the suc- 
cessive directors Dr Alan Hunter and Professor F.Graham 
Smith, the second of whom has done me the honour of 
writing the Foreword to this book (the first having written 
the Introduction to my last). But it is Mr Humphry M. 
Smith, sometime Head of the Time Department - and thus 
'Mr GMT' himself - to whom I owe the greatest debt: with- 
out his unstinting help and advice, my task would have 
been immeasurablv more difficult: he read the technical 
parts of the manuscript, as did his colleagues at Herst- 
monceux Dr Donald Sadler, Mr Philip Laurie, Dr George 
Wilkins, Mr Leslie Morrison, and Dr John Pilkington (who 
wrote Appendix IV). Mr David Calvert took many of the 
photographs. I am grateful also to Mr N. J. Oi lora and Mr 
Andrew Murray. 

Space permits mention of only a very few of the others 
who have contributed so much: Professors Eric Forbes of 
Edinburgh and John North of Groningen, and Mr Beresford 
Hutchinson of the British Museum, all of whom read parts 
of the manuscript; Professors J. Simmons of Leicester, 
Seymour Chapin of Los Angeles, Augusto Salinas of 
Santiago, Chile, Owen Gingerich of Harvard; Colonel 
Humphrev Quill and Commander Edward May of London; 
Mrs Rita Shenton of Middlesex and Mr John Combridge of 
Essex; Mile Suzanne Debarbat, Mme Anna Stoyko, Dr 
Guinot, and M Morando of Paris Observatory; Drs Charles 
Cotter of Cardiff and Barbara Haines of Abervstwvth; Mr 




Norman Robinson and his staff at the Royal Society; Mrs 
Enid Lake at the Royal Astronomical Society; Commander 
Andrew David, Hydrographic Department; and Mrs 
MacNamara at the GPO archives. I am grateful to the 
Marquess of Bute and to his archivist Miss Armet for allow- 
ing me access to the papers of James Stuart Mackenzie, and 
to the Academie des Sciences, Paris, for the photostats 

Finally, I would like to pay tribute to my wife and family 
who have suffered most throughout the writing of this 

National Maritime Museum, Greenwich, 
September 1979 

Derek Howse 

1 Seeking the longitude 300 bc - ad 1675 

Longitude in ancient times 

The concept of geographical latitude and longitude for de- 
fining positions on the Earth's surface had probably come 
into use in ancient Greece before 300 bc, but not in the way 
we think of today - of latitudes so many degrees north or 
south of the equator and longitudes so many degrees east or 
west of some chosen meridian. In I lellenistic times these 
quantities were most often - but not always - thought of in 
terms of time, in the number of hours of daylight on the 
longest day of the year in a particular latitude, in the differ- 
ence in local time between two places for longitude differ- 
ences east or west. So far as we know, the first person to 
provide a mathematically clear theory of geographical lati- 
tude and longitude was Claudius Ptolemy (c. ad 100-165) 
who in his Geography created a consistent grid of co- 
ordinates, reckoned in degrees instead of the traditional 
time co-ordinates, with latitudes measured from the 
equator and longitudes from the westernmost point of the 
known world, the 'Fortunate Isles'. ' 

Ptolemy, who spent most of his working life in the great 
library of Alexandria, produced two major works, the im- 
pact of which continued to be felt until the seventeenth 
century. The first of these was his 'Great Collection' or 
Mcgale Syntaxis, better known as the Almagest, a synthesis of 
all that was best in Hellenistic mathematical astronomy, 
which continued the work of I lipparchos and Apollonios 
and added much material of his own. The second was his 
Geography, a gazetteerand atlas of the known world in eight 
books, giving geographical positions for many thousands 
of places. In a long introductory treatise on map-making 
Ptolemy dealt with the finding of geographical positions, 


discussing in particular the methods advocated three 
hundred years before by Hipparchos of Niicaea in Bithynia 
(c. 190-120 bc), whose proposals for finding longitude are 
pa it icu larly releva n t . 

In I lipparchos's time it was realized that finding differ- 
ences of longitude would be possible if the same event could 
be observed (and the local time of that event measured) in 
each of the two places concerned. Hipparchos suggested 
making use of the eclipses of the Moon for that purpose 
because the entrance of the Moon into the Earth's shadow is 
something which occurs at precisely the same moment for 
all observers regardless of their position on the Earth. What 
Hipparchos failed to explain was how the local time of each 
place should be found. Because the Sun musl be below the 
horizon during a lunar eclipse, a sundial cannot be used 
directly. Various possibilities are discussed in Appendix I. 
But there were other difficulties with the lunar-eclipse 
method. Eclipses are comparatively rare -sometimes two or 
three a year, sometimes none at all - so finding the longi- 
tudes of a large number of places takes a long time. In order 
that no opportunity should be missed, Hipparchos is said to 
have compiled a list of future eclipses for the succeeding six 
hundred years. 

Another difficulty with the lunar-eclipse method is that 
defining a particular point in an eclipse is not easy. The 
'beginning', the 'middle', and the 'end' of an eclipse can 
mean different things to different people, so that errors of 
several minutes of time - and therefore of degrees or more 
of longitude - can creep in from that cause alone. Neverthe- 
less, Hipparchos'a eclipse method was to be the only prac- 
ticable way of finding longitude astronomically for the next 
i(xx> years. 

In his Book I, Chapter IV, entitled 'Carefully observed 
phenomena should be preferred to those derived from the 
accounts of travellers' (one of Hipparchos's tenets three 
hundred years before), Ptolemy discusses Hipparchos's 
lunar-eclipse method of finding longitude: 

. . . and when others coining after him . . . calculating most of their 
distances, especially those which extend to east or west, from a 
certain general tradition, not because of any lack of skill or . . . 
indolence on the part of the writers, but because in their time, the 


use of exact mathematics had not yet been established; and when 
in addition to this not many eclipses of the moon have been 
observed at the same time in different localities as was that edipse 
at Arbela which was noted as occurring thereat the tilth hour, from 
which observation it was ascertained how many equinoctial hours 
[equal as opposed to unequal hours), or by what space of time two 
places were distant from each other east or west; it is just and right 
that a geographer about to write a geography should lay as the 
foundation of his work the phenomena known to him that have 
been obtained by a more careful observation . . . 2 

The eclipse referred to occurred just before the Battle of 
Arbela in 330 bc, recorded at both Arbela and Carthage 
with a supposed difference in local times of three hours. 
(The actual difference should have been 2 h 15 min.) In fact, 
this seems to have been the only astronomically deter- 
mined difference of longitude used by Ptolemy in his 
Geography. 3 

After Ptolemy, more than a thousand years were to pass 
before there were any significant developments. Then, in 
the last few decades of the thirteenth century, there were 
two events of great importance in the story of longitude at 
sea. The first of these was the earliest appearance in Europe 
of the mechanical clock, one of the most important in- 
ventions of the Middle Ages. As the name implies - the 
Latin clocca means bell - the earliest clocks were primarily 
for sounding the hours for religious and secular purposes: 
only later did they become astronomical and navigational 

The second development was the appearance of the sea 
chart, the earliest reference to which is found in an account 
of St. Louis of France during the Second Crusade in 1270. 
Though most scholars are agreed that there must have been 
some form of nautical chart in classical times, the sea chart 
as we know it evolved in Italy and was normally a repre- 
sentation of the Mediterranean drawn in ink on a whole 
skin of vellum, usually to accompany a portolarto, or pilot- 
book (hence such charts came in the nineteenth century to 
be known as 'portolan charts'). 4 The earliest surviving chart 
is of the Mediterranean, unsigned and undated but, from 
internal evidence, seeming to have been drawn about 1300. 
It has no scales of latitude or longitude. 5 


The age of discovery 

At the beginning of the fifteenth century trade between 
Europe and Asia was by way of the Black and Mediterranean 
Seas, or overland, and was almost entirely in the hands of 
the Italian maritime states, Venice and Genoa. Cut off from 
those markets, Portugal looked southward for expansion by 
sea. Since early in the fourteenth century she had enlisted 
the services of Genoese pilots to help create a navy. Under 
the energetic leadership of Prince I lenry 'the Navigator' 
(1394-1460), son of a Portuguese king and an English 
princess, the Portuguese push to the southward was inten- 
sified. Dom Henry was governor of the southern province 
of Algarve and, from his headquarters at Sagres near Cape 
St. Vincent, he employed Arab and Jewish mathematicians 
to instruct his captains in the art of navigation in the 
Atlantic, an art which was to be very different from that 
demanded of seamen in the Mediterranean. 

Eventually, helped by the revenues of the Order of Christ 
of which Dom Henry was Grand Master, the Portuguese 
found a seaway to the East by sailing southward and 
eastward round Africa. Cape Bojador, just south of the 
Canary Islands, was passed in 1434, Cape Verde in 1444, the 
Cape of Good Hope in 1487, and, in culmination of the 
Portuguese efforts, Calicut on the Malabar Coast of India 
was reached by Vasco da Gama in 1498. 

These ocean voyages, this sailing into the unknown, gave 
rise to a new concept in navigation - the use of astronomy 
to supplement the time-honoured methods of compass, 
lead-line, and informed estimation of ship's speed. The 
Portuguese succeeded in devising methods and instru- 
ments to find latitude by observation of the Pole Star and of 
the Sun. This was adequate as long as the ocean voyages 
were primarily in a north-south direction, but, once around 
the Cape of Good Hope, it was east-west distances that 
mattered - and even more so for Christopher Columbus, 
taking a departure from the Canaries in September H9 2 - 

Some explorers did try to find their longitude astronomi- 
cally. To help him do this, Columbus carried with him a 
series of Almanacs or Epliemerides compiled by Johannes 
Mueller (called Regiomontanus after his birthplace Konigs- 
berg) which gave predicted positions of the Sun, Moon, and 

2. The cross-statf in use. both for observing lunar distances and for 
measuring the heights of buildings. From the title page of Peter Apian's 
IntToduCtio Gcograjpiica Petri Apriani in Doctissinms Vcnwri AnnotaKontS . . . 
(Ingolstadt 1533). 

3 The earliest illustration of the lunar-distance method of finding 
longitude. From the first edition of Peter Apian's Cosmographica . . . 
nngolstadtandLandshuti524>. In later editions, the observer has grown a 


planets as seen from the city of Nuremberg for every day 
from 1476 to 1506, 6 later extended to 1531 by Stoffler of 
Tubingen. 7 On his second voyage Columbus observed an 
eclipse of the Moon on 14 September 1494 while at anchor 
off I lispaniola: the longitude he obtained later turned out to 
be some 23° too far to the west, about 1 '/2 hours in time. 8 On 
his fourth voyage, while his ship was aground off Jamaica 
on 29 February 1504, he used a predicted lunar eclipse, first, 
to put the fear of God into the natives (as Mark Twain was to 
do in A Connecticut Yankee in King Arthur'* Court 350 years 
later), and then, by his observations, to find his longitude: 
this time, his error seems to have amounted to more than 
2V2 hours, again too far to the west. (Whether these errors 
arose from his own observations of local time, from errors in 
Regiomontanus's almanac, or whether it was just plain 
fudging to prove that he, Columbus, had indeed reached 
Asia, we may never know. 9 ) Amerigo Vespucci on his 
second voyage observed an occupation of Mars (when the 
planet passes behind the Moon) on 23 August 1499, de- 
riving a longitude of 1366% leagues (82 ) west of the city of 
Cadi/., I0 the accuracy of which is difficult to judge. 

But the methods of Columbus and Vespucci could be 
used very infrequently and then generally only in harbour. 
In the 1490s, therefore, there was a very real need for some 
solution to the problem of finding longitude at sea, a prob- 
lem which was eventually to lead to the founding of 
Greenwich Observatory and the establishment of 
Greenwich time. 

Longitude by lunar distance 

A possible solution to this problem was indeed forth- 
coming. In 1514 Johann Werner of Nuremberg (1468-1522) 
published a new translation of the First Book of Ptolemy's 
Geograyiiy . n In his commentary on Chapter IV (part of 
which is quoted above) Werner put forward a new theory 
for finding longitude which has come to be called lunar- 
distance method, using a cross-staff, an instrument derived 
from the Jacob staff invented some two hundred years 
earlier by the Provencal Hebrew astronomer Levi ben 
Gerson. 12 

Though it took 250 years to become a practical prop- 


Aureus numcrus a Quadragcfima 

zh Fcbruarii 

Cydus folaris 1 Pafca 

\ Aprilis 

Littcrg domkalcs g f Rogationes 

\z AVaii 

lrictino a Afcefio domini 

\c AVaii 

Intuallu a dies Pcntccofte 

zc AVaii 

Septungefima r Februatii Aduit?domi 

1 DcccmbriG 

Eclipfis Lung 

zo 15 3 tf 



1 9C 

Saturnus ~ab initio ani ad c Ma rti i .-item a *i Noucbris ad 
exitum anni retrogradus . 
a caoite anni ad 19 fcbruarii rite a 1% Nouebri* 



ad calccm anni retrogradus. 

Mars a b initio anni ad ^ Fcbr uarii retrogradus. 

Venus abincuntc anno ad zc Ianuarii retrograda. 

M.ercurius a H Aprilis a d zc etufdem : rurfas a v) Iulii 
ad t o Augufii : et a zz Nouebris ad ir 
~T)eccbris retrogradus . 

1. Regiomontanus's almanac, 1474. This page, for 1504, shows the prediction of the total eclipse 
»l the Moon used by Columbus to frighten the natives of Jamaica. 


osition, this method eventually made it possible to measure 
longitude at sea. Explained more fully in Appendix 1, it 
makes use of the fact that the Moon appears to move com- 
paratively quickly against the background of the stars in the 
zodiac belt - approximately its own diameter in one hour. 
'Therefore the geographer goes to one of the given places 
and from there observes, by means of this observational rod 
[the cross-staff] at any known moment, the distance be- 
tween the Moon and one of the fixed stars which diverges 
little or nothing from the ecliptic." 3 So said Werner, who 
was, incidentally, talking of the method being used on land. 
Then, with the aid of astronomical tables for the star's 
position and an almanac for the predicted Moon position, 
he could find his difference of longitude from whatever 
place the almanac was based upon. That, however, was 
only theory. In practice, neither the instruments nor the 
tables were at that time accurate enough to give a useful 
result. Furthermore, Werner omitted to take into account 
lunar parallax - the fact that the Moon appears in a different 
position according to the observer's position on Earth - 
something which is absolutely fundamental if accuracy is 

It was not Werner's own description of the lunar-distance 
method which made it known to seamen and scholars but 
that of Peter Apian (1495-1552), whose Cosmogmphica 
(Ingolstadt, 1524) l4 gave a clearer description as well as 
a picture. Apian's work was re-edited by Gemma Frisius 
(1508-55), the Frisian astronomer and mathematician, in 
1533 1S and successive editions continued to be published 
for the rest of the century. 

The first description of the method in English was given 
by Dr. William Cuningham of Norwich in his Cosmograpliical 
Glasseot 1559, a book distinguished by the beauty of both its 
language and its italic typeface: 

I will shew you, ther are thre thinges required vnto this busines, 
the Astronomers staffe, also called Iacobes Staffe (the makinge of 
which you shall finde among th'other instrumentes) the second is 
the true place of the Mone in the Zodiake, in degrees, & minutes, 
for the hour you make obseruatio, (whiche you may take out of an 
Ephemerides) and the iij. is the longitude of a fixed sterre, which 
you may take out of the Table of fixed sterres in my firste boke. 


These had, you muste take your staffe with the Crosse on it and 
applye the one ende of the Crosse to the Center of the Mone, and 
the other vnto the sterre: which thing to do, you shall remoue the 
Crosse vp and downe, vntill the endes of the staffe touch both the 
center of the mone & also of the sterre. Thys ended, the crosse shall 
shewe vou what the distaunce of the Mone, & starre is in degrees & 
minutes. Then take the distaunce in degrees, & minuts of the 
Mone, & fixed sterre, which you had before the obseruation: And 
substract these ij. distances, th'one out of th'other, the remanet 
deuide by the portio that the mone moueth in one hour, And that 
shall shew you the time, whan as the Mone was ioyned wyth the 
starre (if the starre be West from her) or whan she shall be ioyned 
with the starre, if it be F.ast from the Mone. ' 6 

The chronometer method invented 

The first author known to have proposed the employment 
of a timekeeper for determining longitude at sea is Gemma 
Frisius in his work De Prmcipiis Astronoiniae Cosmographicae 
. . . published in Louvain in 1530. In Chapter 19, 'Concern- 
ing a new method of finding longitude', he says: 

In our age we have seen some small clocks skilfully produced 
which are of some use. These, on account of their small size are no 
burden to a traveller. These will often keep running continuously 
for up to 24 hours. Indeed if you help, they will keep running as if 
with perpetual motion. It is therefore with the help of these clocks 
and by the following methods that longitude is found. In the first 
place we must take care that before we set out on our journey, we 
should observe exactly the time at the place from which we are 
making our journey. Then while we are on our journey we should 
see to it that our clock never stops. When we have completed a 
journey of 15 or 20 miles, it may please us to learn the difference of 
longitude between where we have reached and our place of de- 
parture. We must wait until the hand of our clock exactly touches 
the point of an hour and at the same moment by means of an 
astrolabe or by means of our globe, we must find out the time of the 
place at which we now find ourselves. If this time agrees to the 
minute with the time shown on our watch, it is certain that we are 
still on the same meridian or in the same longitude, and our 
journey has been made towards the south. But if it differs by one 
hour or by a number of minutes, then these should be turned into 
degrees, or minutes of degrees, by the method I set out in the 
previous chapter, and in this way the longitude is discovered. In 
this way I would be able to find the longitude of places, even if I 
was dragged off unawares across a thousand miles, and even 


though the distance of my journey was unknown. But then first of 
all, as always, the latitude must be learnt. I have already explained 
this before and also that it can be found out by various methods of 
finding out the time. Then indeed it must be a very finely made 
clock which does not vary with a change of air. " 

In the 1553 edition a sentence is added which seems to be 
the first actual reference to longitude-finding at sea: 

Therefore it would be useful on long journeys, especially sea- 
journeys, to use large clepsydras (that is water-clocks) or sand 
glasses, which will measure a whole day exactly, through which 
the errors of the other clock may be corrected. "* 

The first description in English was a translation of the 
above by Richard Eden in 1555. He adds his own postscript: 

And so shall the longitude bee founde. And by 

this arte can I fynde the longitude of 

regions althowgh I were a thou = 

sand myles owt of my attemp= 

ted course & in an unkno= 

wen distance, but the 

latitude must furste 

bee perfectly 



- all of which was a pious hope not to be realized for more 
than two centuries. 

Cuningham's charming dissertation on the same subject 
four years later, in the form of a dialogue between the 
4 'interloquutors' Philonicus and Spoudaeus, is reproduced 

in fig. 4. 20 

The longitude prizes 

Though theoretical solutions had been found, the practical 
problem of how to find longitude not only at sea but also on 
land became the more urgent as oceanic voyages became 
more frequent and as nations began to depend more and 
more upon trade with the Indies. 

In Spain in 1567 Philip II offered a reward for the solving 
of the longitude problem at sea. Towards the end of the 
century Miguel de Cervantes and other Spanish authors 
began to make fun of the attempts of some 'crazy people' to 
find position at sea - el punto fijo. 21 'Finding the longitude', 

You/hall prepare k par fait docile artificially madefuch 

as are brought from Flatiders^ we baue the as excels 

lently without Temple barre,made of our countrymen. 

Spoud. T)oyou not meanefucb t as we */e to weare in 

thefacionofaTablet? jPhi. Tea truely } wben asyou 

tr audi joufhallfet the nedle of your e T>iallexaBlye on 

the hour found out by the fonne on the daye^ byfome 

Jlarrein iixnightithetrmelingwitboute intermijjton, 

whan as you haue trauded.xx.yea.xl. miles or more (if 

your next place y wbofe longitude you defire be Jo far di- 


eth;afterwith an JJlrolabe, or Quadrant \finde out the 

hour of the day in thatplace'.O" ifitagre with the fame 

ivhicbyour dockfhewethjbe aff'uredyour place is north 

or South fro the placeyou came from , & therfore haue 

tlxfame logitude, & meridia line. 'But & the time dif 

ferfubtratt til one t out of tb other & the differece turn 

into degrees & minut.ofth'eijuinocliallas before , then 

adde in th' other. ij. precepts ,gomg before. 

'But now behold the Jl^ie is ouercajl with c/oudes; 

wherforc let ms hajle to our lodgings, & 

ende our talkgfor this pre fente. 

ipoud. Witharighte 



A The chronometer method of finding longitude - an English description 
01 '559- From W. Cuntngham, The Cos m ogmpl ri cal Giassf (1559), f. 110. 


along with 'squaring the circle', began to be equated with 
something which was virtually impossible: Cervantes's 
Mathematician in The Dog's Dialogue says: 
I have spent twenty-two years searching for the fixed point [pun to 
fijo[ and here it leaves me, and there I have it, and when it seems 1 
really have it and it cannot possibly escape me, then, when 1 am not 
looking, 1 find myself so far away again that 1 am astonished. The 
same thing happens with squaring the circle, where I have arrived 
so near to the point of discovering it that I do not know and cannot 
imagine how it is that I have not got it in my pocket . . ." 

In 1598 Philip III offered 6,000 ducats as perpetual income, 
plus 2,000 as income for life, and 1,000 for help in regard to 
expenses as a reward to anyone who could 'discover the 
longitude'. The whole prize was never won but consider- 
able sums were disbursed on account (so to speak) to en- 
courage possible inventors. Gould quotes seven instances 
of grants between 1607 an -d 1626, most of which were con- 
cerned with compasses and magnetism. 23 About the same 
date, the States General of Holland offered 30,000 florins. 24 
Portugal and Venice are also said to have offered rewards. 25 

The most famous person to apply for the Spanish prize 
was probably Galileo Galilei (1564-1642), the Italian astron- 
omer. One of his first celestial discoveries with the newly 
invented telescope was that Jupiter had four moons in orbit 
(eight more have been discovered since). With orbital 
periods varying between 1% and 17 days, they appear and 
disappear (as seen from Earth) as they pass behind Jupiter 
itself and their reflected sunlight is eclipsed each time one of 
them passes into Jupiter's shadow. The eclipses always 
occur at precisely the same moment for an observer any- 
where on Earth, and the occultations virtually so. Galileo 
realized that here he had a perfect celestial timekeeper 
which, if the eclipses could be accurately predicted, could 
be used for longitude-finding exactly as Hipparchos had 
suggested using the eclipses of our own Moon - but with 
the added advantage that the eclipses of the former occur 
much more frequently, once or twice a night. 

But there were disadvantages: firstly, one had to use a 
telescope, which it was thought would be difficult (later, it 
proved impossible) to use at sea; secondly, the eclipses were 
not quite instantaneous. However, Galileo, confident that 


these difficulties could be overcome, set about studying the 
motions of the satellites and drawing up tables of the pre- 
dicted times of eclipses. In 1616 he submitted this method 
for the Spanish longitude award. The Spaniards were un- 
impressed and, after a protracted correspondence, Galileo 
seems to have given up the idea of selling it to Spain in 1632. 
In 1636 he tried I lolland, saying he had spent twenty-four 
years perfecting his tables. Unlike Spain, the States General 
were impressed, but negotiations were difficult. Galileo was 
at that stage virtually under house arrest at Arcetri near 
Florence, being closely supervised by the Inquisition who, 
it is said, refused to allow him to accept the gold chain 
awarded him by the Dutch Government. 1 lis death in 1642 
ended the negotiations. 
Seventeenth-century developments 

Galileo made a second contribution to the longitude story - 
and a significant contribution to horology - by his study of 
the pendulum as a method of regulating timekeepers. The 
first recorded use of a weight-driven clock for astro- 
nomical purposes seems to have been in 1484 when Bernard 
Walther, a pupil of Regiomontanus, used such a clock to 
measure the interval between the rising of the planet 
Mercury and the moment of sunrise. 26 Clocks were also part 
of the equipment of the great Danish astronomer Tycho 
Brahc (1546-1601), whose catalogue of 777 stars - enlarged 
to 1,005 an d published by Galileo's contemporary Johann 
Kepler (1571-1630) in his Rudolphine Tables of 1627 27 - was 
still the best available when the Royal Observatory was 
founded in 1675. Tycho purchased and tried four clocks 
between 1577 and 1581 before concluding that the inherent 
defects of sixteenth-century clockwork were too great for 
most astronomical purposes. 

Galileo first drew attention to the value of the pendulum 
as a controller for clocks in 1637, but it was not until the 
last years of his life, 1641-2, that he developed a practical 
mechanism. Because it is controlled by gravity, the pendu- 
lum has a natural, regular motion. It is isochronous, that is, 
it oscillates in equal spaces of time almost irrespective of the 
arc of swing or weight of the bob. The time of swing, 
however, does vary in relation to the pendulum's length, 
and it is therefore easily adjusted. 




Galileo produced designs for a rudimentary pendulum 
clock with an escapement (see Appendix III) to keep the 
pendulum in motion. I (owever, it is believed that no actual 
clock was ever made to his design, so the invention of the 
pendulum clock - as opposed to the discovery of the 
isochronism of the pendulum - is generally attributed to 
the Dutch mathematician, astronomer, and horologist, 
Christiaan Huygens (1629-96). The invention of the pen- 
dulum clock in'1657 and of the balance spring for watches in 
the i670S-in which Huygens also had a hand-were two of 
the most fundamental developments in the science of 


Before returning to Huygens and his timekeepers, we 
must mention a theoretical development in the search for 
longitude. In 1634 Jean-Baptiste Morin, doctor of medicine 
and professor of mathematics at the College Royal in Paris, 
announced that he had discovered the secret of longitude. 
Cardinal Richelieu thereupon set up a commission of an 
admiral and five scholars to examine this claim. An astrol- 
oger and believer in the Ptolemaic Earth-centred universe, 
Morin distrusted clocks and is reported to have said that he 
did not know whether the Devil would succeed in making a 
longitude timekeeper but that it was folly for a man to try. 28 
Basically, he proposed the lunar-distance method already 
outlined, but with the added refinement that he took 
account of lunar parallax, a very important development 
(see p. 8 above). Morin's method was geometrically sound 
but the commission considered he had not 'found the longi- 
tude' because the imperfections of the tables of the stars and 
Moon meant that it could not be used in practice. Morin 
applied to Holland for a reward without success. Event- 
ually, he was granted a pension of 2,000 livres in 1645 by 
Richelieu's successor, Cardinal Mazarin. 

Having invented the pendulum clock in 1657, Huygens 
turned his attention to the longitude problem, convinced 
that the horological approach - to produce a marine time- 
keeper that would keep accurate and regular time for 
months on end in any climate, regardless of the motion of 
the ship - would soon make it possible to discover the 
longitude. He produced various marine timekeepers which 
were tried at sea between 1662 and 1687. In 1668 one of his 


timekeepers, which had kept going during both gales and a 
sea battle, gave a difference of longitude between Toulon 
and Crete as 20 30' as against the true value of 19 13', an 
error of only 100 km or so. 29 His early timekeepers were 
controlled by pendulums but, in anything but a flat calm, 
their going was most erratic. So, in 1674, he abandoned that 
design and proposed to control his marine timekeepers 
with balance and balance spring. Huygens made a brave 
effort but, as we shall see, it was to be a hundred years 
before a satisfactory marine timekeeper could be produced. 
When his longitude proposals had been turned down in 
the 1630s, Morin had suggested that what was needed to 
make them practicable was an observatory to provide the 
necessaiy data. This was achieved with Louis XIV's foun- 
dation of Paris Observatory in 1667, resulting from the 
foundation of the Academie Royale des Sciences the 
previous year, largely at the instigation of Louis's Finance 
Minister Jean-Baptiste Colbert, who was determined to 
make France pre-eminent in science - and at sea. In addi- 
tion to French scientists, distinguished foreigners were 
lured to Paris by the offer of large pensions. Among those 
who accepted were Christiaan Huygens from Leyden, Ole 
Romer from Copenhagen, and Giovanni Domenico Cassini 
from Bologna, all ol whom worked in the observatory with 
Frenchmen such as Jean Picard, Adrian Auzout, and 
Philippe de la Hire. 

The British equivalent of the French Academie des 
Sciences, the Royal Society of London for Improving 
Natural Knowledge, had been founded by Charles II four 
years earlier, in 1662. 'Finding the longitude' was only one 
of the many subjects in natural philosophy which engaged 
the attention of this newly formed learned society. The 
following is verse 26 of a poem entitled 'In Praise of the 
Choyce Company of Philosophers and witts who meete on 
Wednesdays weekly at Gresham Colledge': 

26. The Colledge will the whole world measure. 
Which most impossible conclude. 
And .Navigators make a pleasure 

By finding out the longitude. 
Every Tarpalling shall then with ease 
Sayle any ships to th'Antipodes. 



From internal evidence this poem seems to have been 
written about 1661, a year before the King granted his Royal 
Charter. Of twenty-eight verses altogether, its previous 
twenty-five describe the various other projects considered 
by the embryo society - Wren's Moon globe, Boyle's air 
pump, magnetic experiments with iron filings, how moths 
eat cloth, Evelyn's diving bell, graving and etching, remarks 
on smoke pollution, and Wilkins's Universal Character (an 
artificial language and script). 30 

Two matters concerning the French Academie at this time 
should be mentioned before we embark on the story of 
Greenwich time. One of the first projects undertaken by the 
new academy was the re-mapping of France, with longi- 
tudes being found by observation of Jupiter's satellites, as 
first suggested by Galileo. The re-survey by Cassini and 
Picard proved that the old maps showed France to be larger 
5 than it actually was. The King, displeased by the apparent 

reduction of his territory, is reported to have said that his 
surveyors had lost him more land than his armies had 

The second matter concerned a German called Andre 
Reusner of Neystett who, in 1668, approached Louis XIV 
with a reputedly impeccable solution to the longitude 
problem - an 'odometer', a form of ship's log capable of 
measuring not only speed through the water but, according 
to its inventor, speed over the ground as well. 'He ad- 
dressed himself to the king and obtained a letter of intent by 
which His Majesty, having undertaken to finance a dis- 
covery from which all nations would profit, promised to pay 
the inventor 60,000 livres in one lump sum, and to grant him 
the right to draw 4 sous for every ton of capacity of all 
vessels availing themselves of the discovery. His Majesty 
engaged himself to guarantee this right up to the sum of 
8,000 livres per annum, reserving for himself nothing more 
than the option to withdraw this right on payment of 
100,000 livres. One sole condition attached to these 
magnificent promises: that the inventor demonstrate the 
effectiveness of his invention before M. Colbert, M. du 
Quesne, Lieutenant General in His Majesty's naval forces, 
and MM. Hughuens, Carcavy, Roberval, Picard, & Auzout 

5. France re-surveyed. From Rec. del' Acad, des Sc, To. VII, PI. VII, 430. 

The hairline shows the map of France before the survey. 








^nS Mi.-ln-t 


l.i 1 

JTwtU ^Nantrx 

la FL-chr 




■ tf E R. 

r> e 

T .1.- 

Out J 1 COGKM 

, Ilordi-rtu ; 



Es p^one 






of the Academy of Sciences.' The Commissioners 
appointed by the King duly met and concluded that, al- 
though it was ingeniously devised, it was not capable of 
finding longitude with any certainty. The German was 
required to answer in writing all the problems raised by the 
Academy, and he did so, but not even the 160,000 livres 
already guaranteed could inspire him to find answers ap- 
propriate to annul these objections.' 31 So Reusner never 
got his money. Nevertheless, the story shows how much 
importance the French attached to finding a solution to the 
problem of discovering longitude at sea. As the same 
account says: 'A solution in this matter of longitude would 
be of the greatest possible utility to the public as well as its 
Author; for great rewards are promised whomsoever shall 
first resolve this Problem. Seeing indeed that so many men 
have been at pains to solve the Quadrature of the Circle, 
though that could bring them naught but glory, they cannot 
have been negligent in finding out Longitude, which would 
bring no less glory but incomparably more profit.' 3 - This 
search did indeed occupy the minds of many men, for many 
years to come. 

2 Greenwich time for astronomers 1675-1720 

Setting the scene 

The story of Greenwich time started in England in 1674 
when Charles II had been on the throne for fourteen years. 
England was for the time being at peace, following her 
somewhat unnatural (and domestically very unpopular) 
alliance with France during the Third Dutch War, an 
alliance soon terminated after the Dutch Admiral De 
Ruyter's naval successes had heightened the unpopularity 
of the war in England. 

Three people require introduction, all of whom had a 
great influence - the first quite unwittingly - upon the 
history of astronomy and science generally and upon this 
story in particular. The first of these is Louise de 
Keroualle (1649-1734), a Breton lady who, if she never 
monopolized King Charles It's affections, at least secured a 
greater share of them than any other woman. Daughter of 
Guillaume Penancoet, Sieur de Keroualle in Brittany, 
Louise was maid of honour to the Duchess of Orleans, 
Charles's youngest and favourite sister Henriette-Anne 
('Minette'), who married Louis XIV's brother Philippe in 
1661 and thereby became the second lady of France. 

After the Duchess's untimely death from peritonitis in 
1670, Louise was brought to England by the Duke of 
Buckingham - some say was sent by Louis XIV as a spy - to 
be named maid-of-honour to Queen Catharine. The King 
had grown tired of Lady Castlemaine (pensioned off and 
created Duchess of Cleveland in 1670) so Louise, with her 
dark good looks and baby face, soon established herself as 
'Old Rowley's' mistress, bearing him a son, Charles 
Lennox, later Duke of Richmond, on 29 July 1672. After 
naturalization Louise was created Duchess of Portsmouth 


in 1673 and was henceforward always regarded as the 
King's principal mistress though it has been said that 
'Charles never discarded, he only added to his hand'. ' Nell 
Gwyn of Old Drury, and Hortense, Duchess of Mazarin, 
satisfied another part of Charles's nature: they were com- 
plementary to Portsmouth, never her rival. In the context of 
this story it is interesting to note that Charles, shortly before 
his death, was considering creating Nell Countess of 
Greenwich. 2 

The second is Jonas Moore (1627-79), born at Whitelee in 
Pendle Forest, Lancashire. He resolved early to devote 
himself to mathematics. During the Civil War he had the 
use of the library of the antiquarian Christopher Towneley, 
patron of many young north-country natural philosophers 
such as William Gascoigne, Jeremiah Horrocks, Jeremiah 
Shakerley, and Christopher's own nephew, Richard 
Towneley. Publishing his first work, a mathematical 
textbook, in 1647, Moore came to London soon after, setting 
up as a mathematics teacher. He found all too few pupils 
during those troubled times but was fortunate enough to be 
appointed surveyor in the work of draining the Great Level 
of the Fens from 1649, a post in which he made his reputa- 
tion, being kept busy thereafter with various other surveys 
throughout the Commonwealth period. On the Restoration 
he re-published his Arithiuetick, with a dedication to the 
Duke of York. In 1662 Tangier came to the British Crown as 
part of the dowry of Queen Catharine of Braganza, and 
Moore was sent there in 1663 to survey and report on the 

In 1669 he was knighted and appointed Surveyor- 
General of the Ordnance, henceforward residing in the 
Tower of London where 'he enjoyed high royal favour, 
which he turned to account for rescuing scientific merit 
from neglect', 3 one of the manifestations of which was his 
becoming, in 1670, the patron of the 24-year-old astronomer 
John Flamsteed. Aubrey called Moore 'one of the most 
accomplished gentlemen of his time: a good mathematician, 
and a good fellow', adding, 'Sciatica: he cured it by boiling 
his buttock'. 4 In 1673 Moore was, with Samuel Pepys, one 

6. Louise de Keroualle. From an oil painting bv Henri Gascar, 1670-1, the 
property of Lord Talbot of Malahide. 

7. Sir Jonas Moore. From an engrav- 
ing in his Arithmetkk. 

8. John Flamsteed, first Astronomer 
Roval, with his assistant Thomas 
Weston, sketched from life in 17100V 
Sir James Thornhill, and painted on 
the south-east corner of the ceiling of 
the Painted Hall of Greenwich 
Hospital, today's Royal Naval 


of those responsible for the foundation of the 'Royal 
Mathematical School within Christ's Hospital' to provide 
training in navigation for boys for the King's service at sea. s 

The third character is John Flamsteed (1646-1719), born 8 

near Derby, the son of a rich maltster. At the age of 14 he 
was afflicted by 'a fit of sickness that was followed with a 
consumption, and other distempers', 6 which caused him to 
quit the Free School in Derby two years later and left him 
with a delicate constitution, which was to plague him for the 
rest of his life. It also meant that he could not immediately 
go up to university but had to pursue his studies at home, 
principally in the sciences of mathematics and astronomy. 
By 1669 he was corresponding with Fellows of the Royal 
Society in London, which he visited at Easter 1670. There he 
met Sir Jonas Moore who, recognizing his talents, became 
his patron and presented him with astronomical instru- 
ments - telescope glasses and tubes, and a micrometer - 
with which to pursue his researches in Derby. On the way 
home he visited Cambridge, entering his name for Jesus 
College and meeting, among others, Isaac Newton, with 
whom he was later to have so many disagreements. In 1672 
he visited Richard Towneley at Towneley Hall near Burnley 
in Lancashire to study the observations of three brilliant 
north-country astronomers - William Gascoigne, Jeremiah 
I Iorrocks, and William Crabtree - all of whom had died in or 
before the Civil War, and whose papers had been preserved 
by Richard Towneley's antiquarian uncle, Christopher. 

Thanks to the interest of Sir Jonas Moore, the King issued 
a warrant to Cambridge University on 14 May 1674 'to grant 
an M.A. degree to John Flamsteed, late of Jesus College, 
who has spent many years in the study of the liberal arts and 
sciences, and especially of astronomy, in which he has 
already made such useful observations as are well esteemed 
by persons eminently learned in that science'. 7 

The foundation of Greenwich Observatory 
In the autumn of 1674 the Royal Society began to make 
plans for setting up an observatory in King James I's old 
College at Chelsea (demolished in 1682 to make way for 
Wren's Royal Hospital) which had been presented to the 
Society in 1667. Sir Jonas Moore offered to pay for all 


expenses connected with this observatory and proposed 
that the 28-year-old Flamsteed, who was preparing to take 
Holy Orders the following Easter, should be observer. 

However, while these plans were being thought about 
there occurred an event which was to have a profound effect 
upon our story. In Flamsteed's own words, 'An accident 
happened that hastened, if it did not occasion, the building 
of the [Greenwich) Observatory. A Frenchman, that called 
himself Le Sieur de St. Pierre, having some small skill in 
astronomy, and made [sic] an interest with a French lady, 
then in favour at Court, proposed no less than the discovery 
of the Longitude.' 8 The lady then in favour at Court was 
none other than the Duchess of Portsmouth. We still know 
almost nothing about St. Pierre but, from his name and title, 
it seems quite likely that he came from Brittany and may 
well have had previous links with the Duchess; but research 
has so far failed to identify him. 9 Be that as it may, the King 
was persuaded to issue a warrant on 15 December 1674 
appointing a Royal Commission to examine the Sieur's 

. . . that he [St. Pierre] hath found out the true knowledge of the 
Longitude, and desires to be put on Tryall thereof; Wee having 
taken the Same into Our consideration, and being willing to give 
all fitting encouragement to an Undertaking soe beneficiall to 
the Publick . . . hereby doe constitute and appoint you [the 
Commissioners], or any four of you, to meet together . . . 

And You are to call to your assistance such Persons, as You shall 
think fit: And Our pleasure is that when you have had sufficient 
Tryalls of his Skill in this matter of finding out the true Longitude 
from such observations, as You shall have made and given him, 
that you make Report thereof together with your opinions there- 
upon, how farre it may be Practicable and usefull to the Publick. 10 

The Commissioners were a distinguished group: 

Lord Brouncker, President of the Royal Society, and 

Controller of the Navy. 
Seth Ward, Bishop of Salisbury (Sarum), Savilian 

Professor of Astronomy at Oxford 1649-61. 
Sir Samuel Morland, mathematician, inventor and 

Gentleman of the Bedchamber to Charles II. 
Sir Christopher Wren, the King's Surveyor-General, 

Gresham Professor of Geometry in London 1657-61, 


and Savilian Professorof Astronomy atOxford 1661-73. 
Col. Silius Titus, Gentleman of the Bedchamber. 
Dr. John Pell, mathematician, principally known for 

having invented the sign ■?■ for division. 
Robert Hooke, MA, surveyor, curator to the Royal 

Societv, and Gresham Professorof Geometrv. 

All of these except Wren were already members of a Com- 
mittee appointed by the King the previous year to investi- 
gate the merits of proposals by one Henry Bond for finding 
longitude at sea by measuring the dip of the magnetic 
needle." As for Wren, before he turned to architecture he 
had been an astronomer, so his appointment to the new 
Committee was most apposite. 

On 2 February 1675 John Flamsteed arrived in London to 
stay in the Tower with Sir Jonas Moore. A few days later 
Moore contrived to have Colonel Titus take Flamsteed to the 
King to inform His Majesty of the very important results 
being achieved by Jean Picard and his colleagues of the 
Academie des Sciences in their great survey of France. In 
order to determine longitude Picard was observing the 
eclipses of the moons of the planet Jupiter, a method very 
successful ashore but one which was to prove impracticable 
at sea because the ship's motion made accurate observa- 
tions impossible. In the light of subsequent events it seems 
not improbable that Flamsteed also discussed with the King 
the problem of longitude at sea and the urgent need for an 
observatory in England to provide the basic data needed. 
True, the Royal Society had plans for such an observatory in 
Chelsea, in which Flamsteed was himself involved. But did 
Flamsteed remind the King that Picard had the resources of 
a royal observatory behind him, an observatory - Paris 
Observatory - founded by the French King eight years 

Meanwhile, St. Pierre -and his patron the Duchess- had 
been getting impatient. On 31 January 1675 the Secretary of 
State, Sir Joseph Williamson, told Morland that the Com- 
mittee must set to work immediately. The next day Morland 
sent a letter to Pell (who was acting as secretary) by the hand 
of St. Pierre himself. Eventually, four out of his seven com- 
missioners met at Col. Titus's house on Friday 12 February, 



Morland, Pell, Titus, and Hooke providing the quorum of 
four demanded. At Sir Jonas's instigation (though not 
himself a member), Flamsteed was admitted as an official 
Assistant to the Committee. 

Me [St. I'ierrcl pretended no less than the absolute discover)' of the 
longitude from easy celestial observations, and demanded the 
heights of two stars, and on which side of the meridian they were, 
with heights of the Moon's two limbs [upper and Iower| with the 
pole's height to be given in minutes [i.e. he required the latitude of 
the observer], as also the year and day of observations, whence he 
undertook to show under what meridian these observations were 
made. ' J 

So wrote Flamsteed in 1682. Though not specifically stated, 
it is almost certain St. Pierre was hoping for a large reward 
for finding the solution to this vital problem. 

At the St. Pierre Committee's first meeting in London on 
Friday 12 February, Flamsteed undertook to supply the 
observations called for and these were duly delivered to Pell 
the following Wednesday. Pell passed them on to St. Pierre 
two days later: 

On Friday, Febr.19. I delivered those two observations to that 
Frenchman in the presence of Mr. Payen, a Lorrain Gentleman 
belonging to Secretary Coventry. This Mr. de St. Pierre disliked 
that they were for the years 1672 & 1673. He would have had them 
for the year 1675. I told him that if he could obtain that the King 
should send observers into the Indies, he must tarry till they could 
come back hither again. He said: these were but Calculations, & he 
could doe nothing with them. I answered, that if they were but 
calculations we could easily have calculated for any night of this 
February, &c. 13 

Meanwhile Flamsteed had written down his comments 
on St. Pierre's proposals. Yes, he said, the Frenchman's 
method might work in theory, though better methods were 
known. St. Pierre proposed to use vertical angles to 
measure the Moon's longitude whereas (as can be seen in 
Appendix I) the Moon's movement among the stars is 
basically horizontal. Furthermore, such vertical measure- 
ments were made the more uncertain by atmospheric re- 
fraction (which varied according to the weather) and 
because of imperfect knowledge of the Moon's parallax. 
Flamsteed's 'better method' was, of course, that of lunar 


distances (see Appendix I) where the principal observation 
is more horizontal than vertical. 

But, wrote Flamsteed, whatever method was used, there 
was a fundamental limitation in that, with the then state of 
knowledge, the basic information simply was not available 
to the degree of accuracy needed. For any lunar method one 
must (a) know where the so-called fixed stars are, relative to 
the Sun's annual path (the ecliptic), and (b) be able to 
predict, perhaps years ahead, where the Moon will be rela- 
tive to the stars at the moment the navigator takes his sight. 
From his own observations Flamsteed had proved that the 
best available star catalogue, that of Tycho Brahe, could be 
in error by ten minutes or more, while current tables could 
err by as much as twenty minutes. All of which meant that 
any longitude found could be in error by several hundreds 
of miles. To provide the necessary data, said Flamsteed, 
would demand years of observations with large instru- 
ments fitted with telescopic sights. And so it proved: the 
search for the data demanded by Flamsteed was to occupy 
astronomers the world over for the next 150 years. 

On Wednesday 3 March the Committee - Bishop Sarum, 
Morland, Titus, Pell, and Hooke - met again at Titus's 
house, this time to consider Henry Bond's longitude propo- 
sals and to agree on a report to the King. At that meeting it 
was decided that the report should be presented to the King 
by Sarum, Morland, Titus, and Pell at the Privy Gallerv at 
Whitehall at 8 o'clock the following morning. Hooke was 
furious at being excluded: 'Titus a dog,' he wrote in his diary 
for 3 March, 'I should have been at the King's next morning 
with the report.' On 7 March he added: 'Mr. 1 fill told me of 
Dog Titus his abuse. Query Lord Sarum.' Unfortunately we 
do not know the other side of the story. In the event, only 
the Bishop and Titus entered the King's Bedchamber to 
present the Bond report. But apparently they presented also 
Flamsteed's report on the St. Pierre proposals: 

When Charles II, King of England, was informed . . . [that the basic 
information for finding longitude by lunar observations was not 
available], he said the work must be carried out in royal fashion. He 
certainly did not want his ship-owners and sailors to be deprived of 
any help the Heavens could supply, whereby navigation could be 
made safer.' 4 



And in another account, Flamstc-ed goes on: 

. . . the ingenious gentlemen . . . therefore readily joined with Sir 
Jonas Moore to move the King that an observatory might be built 
and furnished with convenient instruments for making such 
observations as were necessary for correcting the places of the 
Fixed Stars, the Luminaries and Planets, in order to the Discovery 
of the Longitude which was not to be otherways expected, and 
myself to be employed in it, with a salary for my support in the 
work, which I lis Majesty was graciously pleased to grant . . . ,5 

So, without waiting for any of the Frenchman's results, the 
King signed a royal warrant that very day, Thursday 4 
March 1675, appointing John Flamsteed his 'astronomical 
observator', enjoining him 'forthwith to apply himself with 
the most exact care and diligence to the rectifying the tables 
of the motions of the heavens, and the places of the fixed 
stars, so as to find the so-much-desired longitude of places 
for the perfecting the art of navigation'; "■ in other words, to 
provide the observational data so that lunar distances could 
be predicted. The warrant was addressed to the Gentlemen 
of the Ordnance, one of whom was Sir Jonas Moore, 
Surveyor-General of the Ordnance. The Office of Ordnance 
was to pay Flamsteed the far-from-princely salary of £100 
per annum, effective from the Michaelmas before. 

Sir Jonas brought the news to Flamsteed. The latter was 
disappointed with the meagre stipend. Writing in the third 
person in 1710 he says ruefully: 'A larger salary was de- 
signed him at first; but, on his taking orders, it was sunk to 
this.'" He received Holy Orders at the hands of Bishop 
Gunning at Ely Palace in London at Easter 1675. 

The next thing to be thought of was a place to fix in. Several were 
proposed as Hyde Park and Chelsea College. I went to view the 
ruins of this latter and judged it might serve the turn: and better 
because it was near the Court. Sir Jonas rather inclined to I lyde 
Park, but Sir Christopher Wren mentioning Greenwich 1 lill, it was 
resolved on ... I8 

So wrote Flamsteed in 1707. All three sites were Crown 
property, an important consideration in the current state of 
royal finances. There were already plans for cMi observatory 
at Chelsea. Of the site in Hyde Park we know nothing. But 
Wren, the King's Surveyor-General, who had relinquished 


the chair of Astronomy at Oxford only two years earlier, 
chose Greenwich Castle - on high ground overlooking the 
River Thames, in the centre of a royal park, away from the 
smoke of London but reasonably accessible by road and 

The choice must have been quickly made. On the 
Saturday, only two days after the King signed the warrant, 
Hookesaysin his diary: 'At Sir J. Moores. . . . He procured a 
patent for Flamsteed of £100 per annum and an observatory 
in Greenwich Park.' By Sunday the Frenchman had heard 
the news too. He came hotfoot to see Pell without an inter- 
preter. Speaking in Latin, he asked two questions about the 
observations Flamsteed had provided, questions which 
caused Flamsteed to declare subsequently that St. Pierre 
could not possibly know what he was talking about. St. 
Pierre then demanded further observations because he 
declared that the previous ones were fictae el absurdae - 
fictitious and ridiculous. ,9 

Pell did nothing. However, St. Pierre - and presumably 
the Duchess also - continued to importune the King, claim- 
ing that Flamsteed's observations had been invented - 
which, in the strictest sense, was true. 20 On 23 April, seven 
weeks after Flamsteed's appointment, Pell received a 
strongly worded letter from Secretary Williamson: 

23 Apr. 1675 
His Majesty is so daily importuned by Mons. St. Pierre the French 
Longitude Man, that he commands me to signify to you, that 
absolutely to have a finall answer, that is that you forthwith give 
him such Data he pretends are necessary to the work in hand, and 
that we may either have the great service he pretends to, or at least 
a quiet from his further importuning etc. You must please to set 
yourself to this forthwith & not intermit a day till it be finished, and 
an account returned to His Majesty. 21 

Secure in his position and already signing himself As: Re: 
(Royal Astronomer), Flamsteed immediately wrote two 
more scathing reports, one in Latin for St. Pierre, one in 
English for Pell and the King. He said, in effect, that the 
Frenchman did not know what he was talking about and in 
any case had probably borrowed what little good there was 



in his method from Longomontanus and Morin, two long- 
dead astronomers who had submitted solutions to the prob- 
lem many years before. '1 le had no way to come off but by 
pretending that the observations were feigned: I showed 
him that they were not, yet had they been so, they might 
have served for his purpose in some cases; that he had only 
betrayed his own ignorance; and that we knew better 
methods. Upon which, he huffed a little, and disappeared; 
since which time we have heard no further of him.' 22 

Poor St. Pierre! Though he failed to get his hoped-for 
reward - and though his identity is still a mystery - he has 
nevertheless secured for himself a place in the history of 
astronomy by acting unwittingly as a catalyst in the found- 
ation of the Royal Observatory at Greenwich. 

The building of Greenwich Observatory 

Though the choice of site at Greenwich had been quickly 
made, the actual arrangements for the building took rather 
longer. While Flamsteed was waiting he took up residence 
with Sir Jonas in the Tower of London, using the north-east 
turret of the White Tower as a temporary observatory. A 
nice story, probably apocryphal, is told to the tourists by the 
yeomen warders at the Tower today of how Flamsteed 
complained of the nuisance caused by the many ravens then 
in the Tower, presumably because they sat on and fouled 
his telescopes. The King was about to give orders for all the 
ravens to be disposed of when he was informed of a tradi- 
tion which said that, when the ravens left the Tower, the 
Tower would fall, and probably the Throne also. Mindful 
of recent constitutional events, Charles is said to have so 
modified his orders that a limited number of ravens should 
always be kept, which is the reason why one of the yeomen 
warders is still appointed 'raven master'. 

At last, on 22 June 1675, more than three months after 
first discussions, the King sent Sir Thomas Chicheley, 
Master-General of the Ordnance, a warrant authorizing the 

Whereas, in order to the finding out of the longitude of places for 
perfecting navigation and astronomy, we have resolved to build a 
small observatory within Our Park at Greenwich upon the highest 
ground at or near the Place where the Castle stood, with lodging 


rooms for Our Astronomical Observator and Assistant. Our Will 
and Pleasure is that according to such plot and design as shall 
be given you by Our Trusty and well-beloved Sir Christopher 
Wren Knight, Our Surveyor General, of the place and scite of the 
said Observatory, you cause the same to be fenced in, built and 
finished ...-■' 

The castle referred to was on the site of a tower built by 
I lumphrey, Duke of Gloucester (brother to Henry V), soon 
after the park was enclosed in 1437. The tower was rebuilt 
in 1526 and used as a guest house and hunting lodge by 
Henry VIII, about whom a pleasant story is told by George 
Puttenham in 1589: 

As when the King hauing Flamock his standard-bearer with him in 
his barge, passing from Westminster to Greenewich to visit a fayre 
Lady whom the King loued and was lodged in the tower of the 
Parke: the King coming within sight of the tower, and being dis- 
posed to be merry said, Flamock let vs rime: as well as I can, said 
Flamock, if it please your grace. The King began thus: 

Within this tOUJre 
There lieth u flowrt 
That hath my hart 

I la mock for aunswer: Within this bower, she will. &c with the rest in 
so vncleanly termes, as might not now become me by the rule of 
Decorum to vtter writing to so great a Maiestie, but the King tooke 
them in so euill part, as he bid Flamock, auant varlet." 

In Queen Elizabeth's time the tower was sometimes 
called Mirefleur and came to be known as Greenwich 
Castle. When in 1579 the Queen heard that her favourite 
Robert Dudley, Earl of Leicester, had married Lettice 
Knollys the previous year, she 'thereupon grew into such a 
passion, that she commanded Leicester not to stir out of the 
Castle of Greenwich, and intended to have him committed 
to the Tower of London. . . . '" In the event, Leicester was 
not sent to the Tower and was soon back in the Queen's 
favour. During the Civil War the castle was garrisoned by 
Parliament. The date it was demolished is not known: it was 
certainly standing in 1662 when it is seen in a sketch on 
Jonas Moore's map of the Thames, 2b while the King's 
warrant quoted above implies that it was down by June 
1675, which perhaps accounted for the delay in setting the 
building of the observatory in train. 




In the latter part of the warrant the King directed 
Chicheley to give orders to the Treasurer of the Ordnance, 
Sir George Wharton: '. . . for the paying for such materials 
and workmen as shall be used and employed therein, and of 
such moneys as shall come into his hands for old and 
decayed powder which hath or shall be sold by Our Order 
of the first of January last; provided that the whole sum to be 
expended and paid shall not exceed £500. . . .'- 1 The same 
day that the warrant was signed, 22 June, Robert Hooke was 
asked by Wren to 'direct' the building of the observatory. 
'He promised money,' added Hooke hopefully. On 30 June 
Hooke visited the site with Flamsteed and Edmond Halley 
(who was to become the second Astronomer Royal). On 28 
July Hooke went down to Greenwich with Sir Jonas Moore 
and party and 'set out' the observatory, presumably to the 
designs of Wren. In July Flamsteed moved from the 
Tower to apartments in the Queen's I louse at the bottom of 
what is known today as Observatory Hill, in order to be on 
hand to superintend the building. He received an imprest 
for the first £100 for the building on 27 July and himself laid 
the foundation stone on 10 August. 

Economy was the order of the day. To save money, the 
new building was raised on the foundations of the old 
castle, resulting in its walls being i3'/2° away from true 
north - which Flamsteed thought was a pity. Lead, wood, 
and iron came from the demolished Coldharbour gatehouse 
at the Tower; bricks came from Tilbury Fort; two copper 
balls and two great round shot, presumably for the roof- 
turrets and gateposts, came from Ordnance stores; spars for 
the long telescopes came from the Navy Board. 28 And 
labour and other materials were paid for by the sale of 690 
barrels of old and decayed gunpowder from Portsmouth 
and the Tower to Mr. Polycarpus Wharton at 40s. a barrel - 
who presumably made it serviceable again and re-sold it to 
the Ordnance at £4 a barrel.- 9 The whole cost £520. 9s. id., 
just £20. 9s. \d. overspent. 

The roof was on by Christmas. On 29 May 1676 - the 
King's birthday - Flamsteed moved some of his apparatus 
from the Queen's House to the Great Room of the new 
observatory in preparation for a partial solar eclipse on 31 
May which the King said he wished to observe from his new 


Royal Observatory. In the event, the King did not come but 
was represented by Lord Brouncker, President of the Royal 
Society, one of the original St. Pierre committee. 

Flamsteed and his staff of two moved into the house on 10 
July. On 15 September Sir Jonas Moore, Sir John and Lady 
Hoskins, and Robert \ looke went to Greenwich where they 
found Flamsteed and Halley - then friends, though they 
were to quarrel bitterly in later years - putting the finishing 
touches to the new instruments. The following day, 16 
September 1676, the first observations with the great sex- 
tant were recorded: the work of the Greenwich Observatory 
had begun. 

The Great Clocks 

We come at last to Greenwich time proper. As we have 
seen, two of the ingredients in any solution to the problem 
of finding longitudes at sea by the lunar-distance method 
were accurate star catalogues and accurate tables of the 
motion of the Moon: it was specifically to provide these that 
Greenwich Observatory had been founded. A third in- 
gredient was the provision of an accurate instrument with 
which to measure the angular distance between the Moon 
and star or Sun, and to measure the altitudes of these 
heavenly bodies above the horizon: this will be discussed 
briefly in the next chapter. 

But there was a fourth ingredient. Any astronomical 
method of finding differences of longitude must make the 
assumption that, for practical purposes, the Earth rotates on 
its axis at a constant speed. This had always been assumed by 
the Copernicans but it had never been proved. By good 
fortune, at the very time the Royal Observatory was 
founded, the means of proving this had just come to hand - 
the pendulum clock, invented by Huvgens as recently as 
1657. Moore and Flamsteed determined that one of the first 
tasks of the new observatory should be to conduct experi- 
ments to obtain this proof, using clocks more accurate than 
had ever before been made. 

Accordingly, late in 1675, Sir Jonas ordered two clocks for 
the new observatory from Thomas Tompion (1638- 1713), 
the leading clockmaker working in London. These Great 
Clocks, as they came to be called, had several very special 


%m •" 

9 < Ireenwich Observatory from Gloom's Hill, about 1680, with Inigo 
Jones's Queen's House (autre left), now part of the National Maritime 
Museum, the King's House (fark-fl). now part of the Royal Naval 

College. From an oil painting by an unidentified artist, the property 
of the National Maritime Museum. 


features designed to make them the most accurate clocks in 
the world: 

(17) pendulums 13 feet long hung above the movements, 
beating every two seconds rather than the more usual one 
second of the 'royal' pendulum of 39 inches and hung below 
the movements, recently introduced and named for the 
King himself; 

(/») pin-wheel escapements, apparently dead-beat (see 
Appendix III), based partly upon the ideas of Richard 
Towneley, friend of both Moore and Flamsteed; 

(c) very heavy driving weights, so hung that the clock had 
to be wound only once a year. 30 

The two clocks were most unusual and may even have 
influenced Christopher Wren in the design of the Great 
Room, which is the central feature of the building itself, 
today called Flamsteed House. The room was capable of 
accommodating the two clocks, with their long pendulums 
and movements behind the wainscot. This can be seen in 
n the etching, where the dials (with the pendulum bobs 

showing through the little windows above) are to the left of 
the door. In that same picture can also be seen the other 
reason for the lofty and dignified room - the tall windows 
for use with the long telescopes of the day. 

When some six years later Bishop Fell was proposing that 
the new Tom Tower in Christ Church, Oxford, should be 
fitted up as an observatory, Wren tried to discourage him, 
saying that in these days astronomical work was better 
carried out at ground level. 'Wee built indeed an Observa- 
tory at Greenwich not unlike what your Tower will prove, it 
was for the Observator's habitation & a little for Pompe; It is 
the instruments in the Court after the manner 1 have 
described [mural quadrant on a meridian wall, long tele- 
scope suspended from a mast, sextant for angular dis- 
tances] which are used, the roome keeps the Clocks & the 
Instruments that are layed by.'- 11 

In a letter to Richard Towneley on 22 January 1676 
Flamsteed says: 'Our great room being 20 foot high will be 
capable of a long pendulum to be hung above the clock.' 32 
On 6 July the same year he reports: 'We shall have a pair of 
watch clocks down here tomorrow, with pendulums of 13 
foot and pallets partly after your manner, with which I hope 


we may try these experiments much more accurately than 
with the second pendulum which I find, when the wheels 
are a little swarfed with dust, goes much swifter than it 
might, sometimes almost a quarter of a minute per diem.' 33 

The dials of these clocks (which still survive) each have 
the following inscription: 'Motus Annuus (Year Move- 
ment]. S r Jonas Moore Caused this Movement with great 
Care to be Made A° 1676 by Tho. Tompion.' Over two 
months were to pass before they were in going order, but on 
24 September 1676 Flamsteed reported: '3b p.m. Both 
movements set together.' The next day he wrote, 'Sept. 25. 
8.55.00 by the spring hung pend[ulum|: set mine [J.F.'s own 
clock] with it. 8.54.20 circiter by the pivot pendulum.' 34 One 
pendulum was suspended from a spring, the other on some 
form of knife-edge, presumably for experimental reasons. 
Greenwich Mean Time was in use. 

There were plenty of teething troubles, as we might ex- 
pect from clocks of such revolutionary design. The move- 
ments were mounted in the open behind the wainscot so 
they got dirty very quickly (Tompion would not let 
Flamsteed have the key). There was no form of temperature 
compensation on the long pendulums. Dust and swarf, 
strong winds shaking the house, moist weather, cold 
weather, hands sticking, wanting oil - these were the 
various reasons given by Flamsteed for bad going or 
stopping. Tompion had other ideas and blamed Towneley's 
escapement, replacing it with one of his own design. 

Despite these horological troubles Flamsteed was able to 
proceed with his experiments to prove whether or not the 
Earth's rotation was 'isochronical' - whether or not it 
rotated at a constant speed. I lowever, there were complica- 
tions, one of which was caused by the fact that the Sun itself 
is not the most perfect of timekeepers. Because of the tilt of 
the Earth's axis and the fact that its orbit is not exactly 
circular (which means that the Sun is nearer to the Earth in 
the northern winter than in the summer), solar days vary in 
length through the year. The difference from the mean - the 
difference between clock time and sundial time on any day - 
is today called the Equation of Time but was known by to 

Flamsteed as the Equation of Natural Days. He had recently 
published a new and more accurate table of the Equation, 


Jan Feb Mar Apr May J un Jul Aug Sep Oct Nov Dec 

10. The equation of lime. 

the existence of which was known to the Greeks but had 
still, in the seventeenth century, to be accurately quanti- 
fied.' 5 Time by the stars - sidereal time - does not suffer 
from these complications. Therefore Flamsteed erected a 
telescope on the balcony so fixed that the transit of the 
bright star Sirius could be observed day and night through- 
out the year (unless there were clouds). The interval be- 
tween two successive transits gave Flamsteed the precise 
length of one sidereal day, which, with a little arithmetic, he 
could use to check the clocks which were keeping Mean 
Solar Time. 

As can be seen from the following extract from a letter 
from Flamsteed to Towneley dated 12 July 1677, the 
former's troubles were by no means all horological ones: 

I told you in one of mine formerly that our clocks kept so good a 
correspondence with the Heavens that I doubt it not but they 
would prove the revolutions of the Earth to be isochronical, which 
if guaranteed it will follow that the Equation of Days, which I 
demonstrated in the Diatribe is altogether agreeable to the 
Heavens. I can now make it out by three months' continued obser- 
vations, though to prove it fully it is requisite the clocks be per- 
mitted to go a whole year without any alterations. . . . 

If God send me life to see them go a year, I doubt not but I shall be 
able to make out something more by them than is expected, if the 
Office [of Ordnance] starve me not out, for my allowance you 
know is but small and now they are three quarters in my debt. I fear 

1 1 The Great Room, Royal Observatory, Greenwich, about 1676. Froman etching by Trancis Place 
after Robert Thacker. 

12. John Flamsteed, 1684, sketched in an observation book by his assistant Abraham Sharp. From 
RGO MSS. 3'2* v - 


#rp try<r ^z 

n ftfp-if 

04 AS to 

& 1% 

t&aas ^k**p0W*0 X °% 92 o 


Of $5-4fit0$ 21 
IO% 2$ 

(OB 6 &~ 

iH o;. 

81 -44 


I must come down into the country to seek some poor vicarage, 
and then farewell to one experiment. 16 

Flamsteed seems to have been justified in his assertions. 
Reporting to his patron, Jonas Moore, on 7 March 1678, he 
says: 'My theory of the Equation of Days I looked upon but 
as a dream at first because one part on which it was founded, 
viz the isocroneity of the Earth's revolutions, was only 
supposed, not demonstrated by me; but the clocks have 
proved that rational conjecturea very truth. . . .'"So one of 
the ingredients in the solution of the problem of longitude at 
sea had been provided - by means of Greenwich time: it had 
been proved that the Earth rotates on its axis at a constant 
speed. Of course, as we shall see in the last chapter, it has 
now been proved that it does not - quite. But for practical 
navigational purposes, Flamsteed's results were adequate 
and they stood for 250 years. 

Presented to Flamsteed personally by Jonas Moore and 
removed from the observatory by Mrs. Flamsteed in 1720 
after her husband's death, the two Great Clocks still sur- 
vive - one now being in the British Museum, the other at 
Holkham Hall in Norfolk, the property of the Earl of 
Leicester. Their full story and that of Flamsteed's other 
clocks is told elsewhere. ,H The clocks in the Octagon Room 
today (Flamsteed's Great Room) arc copies of the originals. 

Publication of results 

Though not directly connected with time, the rest of 
Flamsteed's doings are very much part of the longitude 
12 story so deserve to be told, if only in brief. 

After the initial foundation, Charles II seems to have 
taken no further interest in his 'observator's' affairs, which 
were directed by Jonas Moore until the latter's death in 
1679. The Board of Ordnance undertook to keep the house 
in repair, pay Flamsteed a salary of £100 a year and another 
£26 for an assistant. Flamsteed had to find his own instru- 
ments and pay for any skilled assistance he might need. To 
make ends meet, he was forced to take in private pupils. 

Flamsteed was the first astronomer systematically to use 
telescopic sights for all his measurements. His earliest major 
instrument was an equatorially mounted sextant of 7 feet 
radius, the frame made by the lower smiths and paid for by 


Jonas Moore who had also provided the two clocks. Be- 
tween 1676 and 1690, 20,000 observations were taken with 
the sextant. In 1684 Flamsteed was presented to the living of 
Burstow in Surrey and in 1688 his father died. His improved 
private circumstances enabled him to commission a new 
fundamental instrument, a 140° mural arc of the same 
radius as the sextant. This instrument, which cost him £120 
and fourteen months' work, was mounted on a meridian 
wall of brick, enabling him to measure directly the co- 
ordinates of the heavenly bodies as they crossed the mer- 
idian. Before he died in 1719 more than 28,000 observations 
had been taken with the mural arc. 

One consequence of the Government's parsimony was 
that Flamsteed regarded his astronomical results as his own 
property, to be published as and when he chose. This was 
the principal cause of the quarrel between Flamsteed on the 
one hand and Newton and Halley on the other, which was 
to bedevil the last twenty-five years of Flamsteed's life. In 
1704 Prince George of Denmark, Consort of Queen Anne, 
agreed to stand the cost of publication of Flamsteed's obser- 
vations, to be supervised by 'referees', including Sir Isaac 
Newton, by now President of the Royal Society. Printing 
proceeded very slowly and stopped altogether on the 
Prince's death in 1708. Relations between Flamsteed and 
the late Prince's referees deteriorated. 

Since the death of Sir Jonas in 1679, Flamsteed had in 
effect been responsible directly to the Monarch, and none of 
them - Charles II, James II, William and Mary, Anne - seem 
to have taken any interest in their Royal Astronomer (the 
term Astronomer Royal was very seldom used at that time). 
In practice, Flamsteed was pretty well his own master. 

On 12 December 1710, however, Queen Anne was per- 
suaded to appoint a Board of Visitors to direct the affairs of 
the observatory, this Board to consist of the President - 
none other than Flamsteed's bite noire Newton - and other 
members of the Council of the Royal Society. Flamsteed was 
very angry and managed to stave off any visitation for the 
time being. 

In 171 1 Queen Anne gave orders that publication should 
proceed. The next year Historiae Coelestis . . . observante 
Johainie Flamstecdio was published in a single volume, edited 



by Edmond Halley, by now Flamsteed's sworn enemy. 
When Flamsteed saw the new book he was furious. All but 
97 sheets had been printed without his having seen them. 
Material purporting to be Flamsteed's own was in fact only 
an abridgement of his results. He was particularly angry 
about the catalogue of star positions which was full of 
errors, and about Halley's preface which was personally 

On Saturday 1 August 1713 Flamsteed had to suffer his 
first (and only) 'visitation' from the Queen's representa- 
tives, who included Newton and Halley. Having given 
them a glass of wine he excused himself, pleading lameness 
(he was 67), saying they could go anywhere in the obser- 
vatory, except for his library. 1 '' 

Just before her death on 1 August 1714, Queen Anne gave 
her Royal Assent to the Longitude Act. In it Flamsteed was 
ex officio appointed one of the Commissioners for Longi- 
tude. However, though one or two hare-brained schemes 
were submitted to him for a professional opinion, there 
seems to have been no formal meeting of the Board of 
Longitude - as the Commissioners became known - during 
his term of office. 

The Queen's death, however, benefited Flamsteed, at 
least in his quarrel with Newton. With the change of 
government from Tory to Whig, Flamsteed once again had 
friends at Court. On 28 March 1716, by King George I's 
order, 300 copies out of the 400 printed of Halley's edition 
of Historiae Coelestis were placed in Flamsteed's hands. 
Thriftily removing the 97 sheets which had his approval, he 
had the rest burned 'as a sacrifice to truth' in April 1716. 40 
But he saved a few copies for those of his friends that were 
'hearty lovers of truth, that you may keep them by you as 
evidences of the malice of godless persons, and of the 
candor and sincerity of the friend that writes to you, and 
conveys them into your hands: for I will not say I make you a 
present of that which is so odious of itself, and will be 
detested by every ingenuous man.' 41 

These were strong words, but were they really justified? 
It is difficult to decide who was in the right in these rather 

13. Flamsteed's starcatalogue, from his Hbtoria Coeleslh Britaniiica, vol. III. 
published posthumously in 1725. 

t. ,>»■: Zv :*lf 




Ad Annum Chnfti Completum, 1689. 

Ab ObfcrvationibusGRENOviciinOnsERVATORio Regio habitis. 

Afliduis Vigilijs, Cum, & Studio 
Dcductus 6c Supputatus. 


In Conftclhrionc ARII-.TIS. 

Stellar u m 



d>i* in Cornu duirum prcioicnt 
Sequent & Btrti trt 

In Ccrvict 

In VtHiir 


Tnfor.l'up.C.iput,/. ■•i.l.i !'.'« 

InRoflro >hi.<:.ini Hero 

10 46 

•I 'S 45 

Jl jl 15 

31 ?l I? 

.4 8 ,0 

I . 


.i Polo fi. 

! 7 

*9 17 <5 
71 15 35 
74 'O 55 
74 3* 55 
7» 14 45 
o 4? 55 

'4 39 
'5 7 c 
:? 11 c 

J6 3» 9 

17 19 
'7 '9 


37 J< 

38 33 

s8 '4 45 
18 51 
a" 53 45 


57 45 

7? 43 15 
«7 5« >5 

*5 35 
5 43 15 

3068 1 45 
3C65 31 15 
3C71 <8 4< 

*5 3' 4' 
ft >(• 35 

-1 i; i' 

T36 58 

36 48 

36 3.6 



36 49 
38 M 
Jj ;- 

a c j4 

T>9 in 
% • 33 

3 >« 

4 » 
J <5 

3 >« 

4 4° 

: :3 



5 4 



1 4 

9 ' 
5 5" 

5 iB 
:6 E 

5 =3 

7 'A 

8 33 

5 3« 

<5|io 47 

59 H 


I : . 

1; i 

1 3 I! 


J8 7 

57 J> 

■' ■ 


i: :>. 

.1 31 
' J 4 
9 57 
13 5 
5 5* 


: B 

1: 11 
53 u 
18 E 

35" 57 

7 33 

4 R 

c I! 




I j. j P. 

11 - - • 

ft >: 

57 54 

59 4S 

- 1 

59 - 

s8 ; 

6; 4 
5J 1 

3 ST. I 

1: 4 



3 6 

I3 6 

o<J| 8 

o: 6 

«■• 7 


tedious quarrels. Though there seems to have been some 
justification for Fiamsteed's attitude towards the scientific 
establishment in England, he, as a civil servant, had a duty 
to co-operate with others - or would have by today's 
standards. Many of the wrongs he imputed were imagined 
ones and his prickliness, due at least in part to his chronic ill 
health, must have made any dealings with him frustrating 
in the extreme. 

After the publication of what came to be called Halley's 
'pirate' edition of 1712, Flamsteed resolved to print his 
observations at his own expense. Before he died on 31 
December 1719, all of Volume I - observations 1669-88 
including the 97 sheets he had saved from the fire of 1716- 
and most of Volume II - observations 1689-1719 - had been 
printed. The completion of Volume II and the whole of 
Volume III - containing a Latin preface 42 and his British 
13 Catalogue of star positions (thereby fulfilling King Charles's 

1675 directive 'to the rectifying ... the places of the fixed 
stars') - was undertaken by his two former assistants 
Abraham Sharp and Joseph Crosthwait who, incidentally, 
never got paid for their work. 

The three volumes of Fiamsteed's Hisloria Codestis 
Britminica were eventually published in 1725, followed in 
1729 by his Atlas Coelestis, a star atlas which, in effect, put his 
British Catalogue in graphic form. These were indeed 
monuments to fifty years' labour by a great astronomer. 

J Greenwich time for navigators 1700-1840 

The Longitude Act 

On 29 September 1707 Admiral Sir Clowdisley Shovel with 
twenty-one ships of the Royal Navy sailed from Gibraltar 
for England. Britain was at war with France and, though 
Gibraltar had fallen into British hands in 1702, it was con- 
sidered unwise for the larger ships of the fleet to remain in 
the Mediterranean during the winter. Weather on passage 
was not good. There were westerly gales from the 5th to the 
10th of October. The 12th and 13th were squally, followed 
by two days of light winds. On the 16th and 17th there were 
easterly gales and on the 19th north-westerly gales. On the 
21st the sky cleared and several ships got observations for 
latitude, while sounding gave a depth of between 90 and 
140 fathoms, showing that the fleet was on the edge of the 
Continental Shelf. Three ships having been detached to 
Falmouth for convoy duty on the 22nd, and no sun being 
visible all day, the remaining 18 ships hove to in the after- 
noon to obtain soundings. Then, satisfied that they were in 
the mouth of the English Channel and clear of all danger, 
the ships ran to the eastward before a favourable gale. 

About 7.30 p.m. the same day the Association, Eagle, and 
Romney struck the Gilstone Ledges in the Scilly Isles. Of the 
1,200 men in the three ships, only one man was saved - the 
quartermaster of the Romneij. Sir Clowdisley's body was 
found floating the following day and now lies in one of the 
largest and ugliest tombs in Westminster Abbey. The St. 
George and Firebrand also struck the rocks but got off. The 
latter, however, was badly holed and sank later, leaving 
only twenty-three survivors. ' 




I ' . 



s^e ■. 

14. The wreck of Sir Clowdislev 
Shovel's fleet, 1707. Line 
engraving by an unidentified 


15. William Whiston. From an 
oil painting by an unidentified 
artist, the property of Sir John 
Conant, Bart. 


This disaster - four ships were lost, with nearly two 
thousand men - was a profound shock to the British public. 
There had recently been other disasters: in 1691 several 
men-of-war off Plymouth were wrecked through mistaking 
the Deadman for Berry Head; in 1694 Admiral Wheeler's 
squadron, leaving the Mediterranean, ran aground on 
Gibraltar when they thought they had passed the Strait. 
And there were soon to be more. In 1711 several transports 
were lost near the St. Lawrence river, having erred 15 
leagues in their reckoning during twenty-four hours; and 
011722 Lord Belhaven was lost on the Lizard the same day 
on which he had sailed from Plymouth. 2 

A recent analysis of the Clowdisley Shovel disaster con- 
cluded that, while uncertain currents, fog, and bad com- 
passes might have contributed to the disaster, much more 
must be attributed to the lack of accurate knowledge of the 
geographical positions of the headlands concerned, to bad 
charts, to bad navigation textbooks, and to the generally 
low standard of accuracy of navigational practice at that 
time. 5 Though this disaster was not actually caused by the 
lack of a method of finding longitude at sea, nevertheless its 
very magnitude made such an impression on the British 
public that they became more than ever receptive to any 
suggestion that might make navigation safer - and in the 
1710s 'finding the longitude' Beemed to hold the key to this. 
King Charles's foundation of the Royal Observatory was 
obviously not enough (probably less than one in twenty of 
the population knew about that anyway). Something more 
must be done by the Government of his niece. Queen Anne, 
then in the throes of a ministerial crisis. 

On 14 July 1713 there appeared a letter in the journal Tin- 
Guardian, introduced by the editor Joseph Addison with 
these words: '. . . It [the letter] is on no less a subject than 
that of discovering the longitude, and deserves a much 
higher name than that of a Project, if our language afforded 
any such term. But all 1 can say on this subject will be 
superfluous when the reader sees the names of those 
persons by whom this letter is subscribed, and who have 

done me the honour to send it to me '* The writers of the 

letter which followed were William Whiston (1667-1752) 15 

and 1 lumphrey Ditton (1675-1714), Mathematical Master at 



Christ's Hospital. In 1703, after several years as a country 
vicar, Whiston had succeeded Sir Isaac Newton as Lucasian 
Professor of Mathematics at Cambridge, but his theological 
views led to his expulsion from the university in 1710. 
Whiston then set up as a lecturer on scientific and religious 
subjects, the first being a series on astronomy given at 
Button's Coffee House with the encouragement of Addison 
and Steele. 5 The letter started: 

It is well known, Sir, to yourself and to the learned, and trading, 
and sailing worlds that the great delect of the art of navigation is 
thai a ship at sea has no certain method in either her eastern or 
western voyages, or even in her less distance sailing from the 
coasts to know her longitude, or how much she has done eastward 
or westward, as it can easily be known in any clear day or night, 
how much she is gone northward 01 southward.'' 

Having surveyed possible methods of finding longitude, 
Whiston and Ditton went on to say that they had a method 
ol their own, one which would provide not only longitude 
but latitude as well, which they would disclose if they were 
offered a reward, subject to a satisfactory report by Sir Isaac 
Newton and other persons. 

The method eventually disclosed, in a book published in 
1714, 7 turned out to be a proposal that vessels should be 
moored in known positions at intervals along the trade 
routes, each fitted with a mortar which would, every mid- 
night by Peak of Tenerife time (Whiston and Ditton's pro- 
posed prime meridian), fire vertically a projectile, some 
sort of tracer or rocket visible from afar, which would cul- 
minate -• or perhaps burst - at 6,440 feet preciselv. Ships 
would look out for these projectiles at midnight. A compass 
bearing would give the direction from the recorded position 
of the lightship. The distance of the viewer from the light- 
ship could be obtained either by noting the difference in 
time between seeing the flash of the gun's discharge and 
hearing the report, or by measuring the elevation of the 
highest point of the shell. This proposal was still more 
fantastic because the light-vessels in deep water were to be 
moored by lowering some form of sea-anchor through the 
upper layers of water to the allegedly immovable layers 

Bizarre though this scheme may now appear, it seems in 


general to have been taken seriously by some of Whiston 
and Ditton's contemporaries. Not so the Scriblerus Club, 
however, that club devoted to satirizing false learning. 
The group of wits comprising the club - Dr. Arbuthnot, 
Alexander Pope, John Gay, Jonathan Swift, Thomas 
Parneli, and Queen Anne's Lord Treasurer, the Earl of 
Oxford - were convulsed by the publication of Whiston and 
Ditton's long-awaited book. Arbuthnot wrote to Swift on 17 
Julv 1714, a few days after the book appeared: 'Whiston has 
at last published his project on the longitude; the most 
ridiculous thing that was ever thought on. But a pox on him! 
He has spoiled one of my papers of Scriblerus, which was a 
proposal for the longitude not very unlike his . . . that all the 
Princes of Europe join to build two prodigious poles, upon 
high mountains, with a vast lighthouse to serve for a pole 
star.' 8 Some time afterwards, a set of unsavoury verses 
began to circulate: 

Ode. for Mustek. On the Lotigitude 


The Longitude mist on 

Bv wicked Will Whiston 
And not better hit on 

Bv good Matter Ditton. 

So Ditton and Whiston 

May both be bep-st on; 
And Whiston and Ditton 
May both be besh-t on. 

Sing Ditton, 
Besh-t on; 
And Whiston 
Bep-st on. 

Sing Ditton and Whiston, 

And Whiston and Ditton, 
Besh-t and Bep-st on, 

Bep-st and Besh-t on. 

Da Capo 9 

The authorship of these verses has long been the subject 
of debate in bibliographical circles. Published in 1727 in 



the 'Last' Volume of Pope's and Swift's Miscellanies, they 
were ascribed first to Pope, then to Swift. Joseph Spence 
attributed the verses to Cay in 1820. I lowever, the ques- 
tion seems to be settled by the following extract from a 
letter dated 23 December 1714 from Sir Richard Cox, former 
Lord Chancellor of Ireland, in Dublin, to Edward South- 
well, Secretary of State in Whitehall, London: 'Arch- 
deacon Parnell has made the following dirlv lines, which 
are valued, because they Ridicule the Confident Arrian 
Whiston (Whiston had a reputation for holding Arian be- 
liefs]. '"' Then follow, under the title 'A Round O on the 
Longitude by Whiston & Ditton', a text slightly shorter 
than, and varying in details from, the text quoted above. 

Whiston ,\\id Ditton were nevertheless encouraged bv the 
feelings of the public at large on the subject of longitude and 
presented a petition to Parliament at the end of April 1714, 
begging for a reward for 'discovering the longitude' and 
submitting their own scheme. The I louse of Commons 
Journal tor 25 May reprinted another petition a month later: 

A Petition of several Captains of her Majesty's Ships, Merchants of 
London, and Commanders of Merchantmen, in behalf of them- 
selves, and all others concerned in the Navigation of Great Britain, 
was presented to the House, and read; setting forth, ITiat the 
Discovery of the I ongitude is of such Consequence to Great Britain, 
for Safety ol the Navy, and Merchant Ships, as well as Improve- 
ment of Trade, that, for want thereof, many Ships have been 
retarded in their Voyages, and many lost; but if due Encourage- 
ment were proposed by the Publick, lor SUCfl as shall discover the 
Same, some Persons would offer themselves to prove the same, 
before the most proper Judges, in order to their entire Satisfaction, 
for the Safety of Mens lives, her Majesty's Navy, the Increase ol 
Trade, and the Shipping of these Island-, and the lasting I lonourof 
the British Nation: And praying their Petition may be taken into 

This petition was sent to a Committee of the House who 
in turn CO-opted technical experts, including Sir Isaac 
Newton, President of the Royal Society, and Edmond 
I (alley who later succeeded Flamsteed as Astronomer 
Royal. Sir Isaac gave a summary of the current position, 
reported thus to the I louse on 1 1 June: 

Sir Isaac \cuion, attending the Committee, said. That, for deter- 


mining the Longitude at Sea, there have been several Projects, true 
in the Theory, but difficult to execute: 

One is, by a Watch to keep I ime exactly: But, by reason of the 
Motion of a Ship, the Variation of Heat and Cold, Wet and Dry, and 
the difference of Gravity in different Latitudes, such a Watch hath 
not vet been made: 

Another is, by the Lclipses of lupiter's Satellite*: But, by reason of 
the Length of Telescopes requisite to observe them, and the Motion 
of a Ship at Sea, those Fclipses cannot yet be there observed: 

A Third is, by the Place of the Moon: But her Theory is not yet 
exact enough for this Purpose: It is exact enough to determine her 
Longitude within Two or Three Degrees, but not within a Degree: 

A Fourth is Mr. Ditton'* Project: And this is rather for keeping an 
Account of the Longitude at Sea, than for finding it, if at any time it 
should be lost, as it may easily be in cloudy Weather: I low far this is 
practicable, and with what Charge, they that are skilled in Sea- 
affairs are best able to judge: In sailing by this Method, whenever 
they are to pass over very deep Seas, they must sail due East or 
West, without varying their Latitude; <\\\l\ if their Way over such a 
Sea doth not lie due Last, or West, they must first sail into the 
Latitude of the next Place to which they are going beyond it; and 
then keep due Last, or West, till they come at that Place: 

In the three first Ways there must be a Watch regulated by a 
Spring, and rectified every visible Sun-rise and Sun-set, to tell the 
Hour of the Day, or Night: In the Fourth Way, such a Watch is not 
necessary: In the first Way, there must be Two Watches; this, and 
the other mentioned above: 12 

5 1 

The House accepted the report unanimously and directed 
that a Bill should be prepared by several Members, among 
whom were Joseph Addison and General James Stanhope, 
later the first Farl. A Hill for Providing a Publick Reward for such 
Person or Persons as shall Discover the Longitude at Sea was 
presented to the Commons on 16 June. It offered rewards of 
unprecedented magnitude to '. . . the First Author or 
Authors, Discoverer or Discoverers of any such Method . . . 
To a Reward, or Sum of Ten Thousand Pounds, if it Deter- 
mines the said Longitude to One Degree of a great Circle, or 
Sixty Geographical Miles;* to Fifteen Thousand Pounds, if it 
Determines the same to Two Thirds of that Distance; and to 

*A geographical mile is the distance subtended by .in arc-minute on the 
equator, or 6,087 feet, very close to the nautical mile whose length, how- 
ever, varies with latitude because the Earth is not a perfect sphere. Hie 
English statute mile is 5,280 feet. 



Twenty Thousand Pounds, if it Determines the same to One 
half of the same Distance . . /** A half of the reward was to 
be paid as soon as the Commissioners were satisfied that 
'any such Method extends to the Security of Ships within 
Eighty Geographical Miles of the shores, which are the 
Places of greatest Danger', the other half to be paid 'when a 
Ship by the Appointment of the said Commissioners, or the 
major part of them, shall thereby actually sail over the 
Ocean, from Great Britain to any such Port in the West- 
Indies, as those Commissioners, or the major part of them, 
shall Choose or Nominate for the Experiment, without 
losing their Longitude beyond the limits before mentioned'. 

In terms of the 1980s, £20,000 is perhaps equivalent to 
nearly half a million pounds. A prize indeed! But the Bill 
went on to stipulate that, before the reward could be paid, 
'. . . such Method for the Discovery of the said Longitude 
shall have been tried and found Practicable and Useful at 
Sea . . .'. The definition of the phrase 'practicable and 
useful' was to cause much acrimony in future years. 

I here were two further provisions of significance in the 
Bill: first, that the Commissioners were empowered to 
advance sums up to £2,000 for promising schemes, 'to make 
experiment thereof; secondly, if a proposal on trial did not 
quite match up to the specifications for the main rewards, 
but was nevertheless found to be 'of considerable Use to the 
Publick', then it could qualify lor some lesser reward, at the 
Commissioners' discretion. The provisions of the Bill 
applied to all who satisfied the conditions, regardless of 

I he Bill was read for the first time on 17 June 1714, passed 
by the Commons on 3 July, and passed by the House of 
Lords on 8 July. Queen Anne gave her Royal Assent on 20 
July, only twelve days before her death. 

The Commissioners appointed by the Act, who came to 
be known as the Board of Longitude, were the Lord High 
Admiral of Great Britain or First Commissioner of the 
Admiralty; the Speaker of the 1 louse of Commons; the First 
Commissioner of the Navy; the First Commissioner of 
Trade; the Admirals of the Red, White, and Blue Squadrons; 
the Master of Trinity House; the President of the Royal 
Society; the Astronomer Royal; the Savilian, Lucasian, and 


Plumian Professors of Mathematics in Oxford and Cam- 
bridge; and ten named Members of Parliament. 

The immediate effect of the passing of the Longitude Act 
was to stimulate the publication of many pamphlets by 
those who thought they had solutions to the longitude 
problem, the majority of which seem to have had to do with 
timekeepers of one sort or another; several are quoted by 
Gould. 14 Two proposals have survived among the papers 
of John Flamsteed who, as we have seen, was one of the 
Commissioners appointed by the Act: the first was by 
William Hobbs, 'c/o Mr. Jam. Hubert's watchmaker, in 
Finch Lane near the Royal Exchange', for his 'Movement 
with a Decimal Horloge'; ls another by Digby Bull of Three 
Logs Court in White Cross Street, London, for his 'ship- 
watch'. 16 Both asked for trials to be carried out but neither 
received any recognition. 

Even the eminent Sir Christopher Wren, then aged 82, 
put in a claim. The following paper was found in the nine- 
teenth century among the Newton manuscripts at the Royal 


Sir Christopher Wrenn's Cypher, describing three Instruments 
proper for discovering the Longitude at Sea, delivered to the 
Society Novemb. 30, 1714, by Mr. Wren [the son]: 


Vera Copia - Edm: I lalley" 

This was deciphered by Francis Williams of Chigwell, Essex, 
and reported at the British Association meeting of 1859: 

(n) Reverse each line end for end. 

(/>) Strike out every 3rd letter in each line: these give names and 

(c) The residue is the text, thus: 

yd letters 




wovnd in yacvo = Watch - 
magnetic balance 
wound in vacuo 






= Fix head, hips, hands - 
poise lube on eyes 


= Pipe SCiew moving 
wheels from beak 18 

A detailed explanation of this cryptogram has never been 
found but the first phrase must refer to what would be 
called today a marine chronometer with a magnetic balance, 
the whole being kept in a vacuum, a measure advocated by 
Wren (erroneously, as it turned out) many years before to 
combat the effects of temperature changes on the time- 
keeping. The second phrase seems to refer to Wren's ideas 
for observing Jupiter's satellites at sea, the main difficulty of 
which arises from the fact that the motion of the ship makes 
it impossible (as it has since been proved) to hold the tele- 
scope steady enough for accurate observations. The third 
must refer to some kind of log for measuring a ship's speed 
through the water, not unlike that proposed by Reusner to 
Louis XIV in 1668. ,9 

The cryptogram was a device commonly used in the best 
scientific circles of the time to establish priority of invention 
or discovery without actually disclosing any thing that might 
be seized upon by a zealous colleague. It was used by 
Galileo, Huygens, and Hooke, the last of whom published a 
Latin crvptogram giving (when translated) the principle of 
the balance spring in a watch: 'As the tension is, so is the 
force' (that is, the force exerted by the spring is proportional 
to the amount of tension). 

None of these early proposals came to anything. If the 
Board of Longitude met to consider any of them, no record 
has survived, the earliest meeting for which minutes are 
available being 30 June 1737, twenty-three years after its 
establishment. Meanwhile, exactly as had happened a little 
over a hundred years before in Spain, the phrase 'finding 
the longitude', coupled once again with squaring the circle 
(the Board received many suggestions for doing this, all 

i(S. The longitude lunatic, detail from 1 logarth's The Kiike't Progress, plate S, 
1st state, 1735. 


impracticable and none really concerned with longitude), 
passed into the English language as expressing something 
which, if not downright impossible, was extremely difficult 
to achieve. It was used as a catch-phrase in newspapers and 
broadsheets. Gulliver (published 1726), describing what he 
would do if he became immortal like theStruldbrugs, says, 'I 
shall then see the discovery of the longitude, the perpetual 
motion, the universal medicine, and many other great in- 
ventions brought to the utmost perfection.' 20 In 1735, in the 
final, madhouse, scene in his series of paintings The Rote's 
Progress, Hogarth includes a man scribbling on the wall 
calculating the longitude, together with the religious 
maniac, the mad tailor, the mad astronomer, the mad 
musician, the man who fancies himself Pope, and other 
16 lunatics. And on the wall is a sketch of a ship with a mortar 

discharging a bomb vertically, a reference to the Whiston- 
Ditton proposal of 1714. 2I Even as late as 1773 Goldsmith 
makes the ingenuous Marlow say: 'Zounds, man! we 
could as soon find the longitude!' in reply to Tony 
Lumpkin's complicated (and fictitious) directions for 
finding Hardcastle's house. 22 

French prizes 

In France the State did not follow the British example. 
I Iowever, on 12 March 1714 (two months before the longi- 
tude debates in the British House of Commons) Rouille de 
Meslay, French parliamentary counsel, drew up a will in 
which he bequeathed 125,000 livres (at about 20 livres to the 
pound sterling) to found two prizes, to be awarded 
annually by the French Academie des Sciences for philo- 
sophical dissertations on two specific subjects of their 
choice. The larger prize was to be for a dissertation on the 
chosen subject concerning the make-up and motions of the 
solar system and of the principles of light and motion; the 
smaller was to go to him 'who best achieved the shortest 
and easiest method and rule for taking the heights and 
degrees of longitude at sea exactly, and [who made] useful 
discoveries for navigation and great voyages'. 23 Rouille 
died in 1715 and the Academie accepted the bequest in 
March 1716. Despite attempts by his son to get the will 
annulled, the legacy was confirmed by the High Court in 


1718. In the meantime another prize had been offered in 
1716 bv Philippe, Duke of Orleans and Regent of France, to 
be awarded by the Academie (the amount unspecified) to 
the inventor of the secret of finding the longitude. For 
reasons that are not known, the Academie never awarded 
such a prize. 24 

The first Rouille prize was offered in 1720 for the solution 
to the problem of finding the best way to ensure the even 
going of a clock at sea, whether by the design of the move- 
ment itself, or by the suspension. The prize of 500 livres was 
won by a Dutch horologist called Massy for his proposals, 
on paper only, for a watch to be kept in a box maintained at 
an even temperature by a lamp. 25 In fact Rouille's capital 
sum proved insufficient to allow the prizes to be awarded 
annually so, for the next seventy years, a single prize of 
2,000 livres was offered every two years or more, as funds 
permitted. The second prize in 1725 today seems somewhat 
bizarre: 2,000 livres to Daniel Bernoulli, the Swiss scientist, 
for his answer to the question of how best to keep even 
movement at sea of clepsydras (waterclocks) and sand- 
glasses. Later prize questions concerned compasses, the 
motion of the Moon, measurement of altitude, the best 
height of masts; while others concerned marine time- 
keepers. 2( ' 

The invention of the sextant 

Before returning to the longitude problems as such we must 
take note of the development of angle-measuring instru- 
ments because, as is explained in Appendix I, the lunar- 
distance method of finding longitude at sea required the 
navigator to measure very precisely both the angle between 
the Moon and the Sun or star, and the altitude of both 
bodies above the horizon. Werner in 1514 had proposed the 
use of a cross-staff for this purpose. I Iowever, astronomers 
soon realized that such an instrument would never have the 
necessary accuracy. Assuming a lunar distance could be 
measured to an accuracy of half a degree (which with a 
cross-staff it probably could not), then this would be equi- 
valent to an uncertainty of one hour in finding Greenwich 
Time, or of 15 in longitude. But what was needed was to 
find longitude to i° or better, which demanded a lunar- 



distance-measurement accuracy of at least two minutes of 

The first new idea of any promise was reported to the 
Royal Society in London by Robert flooke who, early in 
1666, announced that he was preparing 'a perspective . . . 
for observing positions and distances of fixed stars from the 
Moon by reflection', explained by him with drawings later 
the same year. In 1691 llalley produced designs for an 
instrument which was very much the same, so much so that 
Halley, under pressure from Hooke, withdrew and 
admitted the latter's priority of invention. 27 There is no 
record of either of these instruments having been tried at 

Then came two other developments, one abortive and 
one fruitful. In 1729 the Academie offered the third Prix 
Rouille for the best methods of observing the altitude of the 
Sun and stars at sea, by instruments known or to be known. 
An award of 2,000 livres was made to Pierre Bouguer, pro- 
fessor of hydrography at Croissic, for his design of a back- 
staff in the form of a quadrant where the observer could see 
both Sun and horizon simultaneously. 28 However, as it 
worked on a shadow principle, it could not be used for lunar 
distances. The second event was no less than the simul- 
taneous invention in 1731 of a double-reflection quadrant, 
the ancestor of today's sextant, by two people quite in- 
dependently on both sides of the Atlantic. John Hadley 
(1682-1744), vice-president of the Royal Society who had 
already made many improvements to the reflecting tele- 
scope, produced designs for two instruments on a principle 
which was thought to be entirely new, that of double reflec- 
tion, the ray of light from the heavenly body reaching the 
eye after being reflected twice, once by a mirror attached to 
a moving index-arm, then by a mirror fixed in relation to 
the sighting vane. This allowed both bodies (for lunar dis- 
tances) or the body and the horizon (for altitudes) to be seen 
by the observer simultaneously, making observations in a 
moving ship practicable. Hadley described these instru- 
ments in a communication to the Society in May 1731 and 
they were tested at sea in the Chatham yacht in the Thames 
estuary from 31 August to 1 September 1732, the errors of 
altitudes and of the distances between two stars obtained bv 


the second of the two instruments he had described proving 
to be less that 2 arc-minutes, well within the requirement for 
lunar distances.- 9 

Then, in 1732, the Royal Society received a letter through 
Hdmond Halley from James Logan, Chief Justice of 
Pennsylvania, enclosing affidavits purporting to prove that 
Thomas Godfrey (1704-49), a glazier and self-taught 
astronomer from Philadelphia, had designed a double- 
reflection instrument capable of observing lunar distances 
substantially the same as the first of the two designed by 
Hadley. When Benjamin Franklin started in business in 
Philadelphia with I lugh Meredith he sub-let part of his 
house to Godfrey, who became one of the founder- 
members of the Leathern-apron Club, or Junto, a debating 
club founded by Franklin, devoted to morals, politics, and 
natural philosophy. Franklin, though he called Godfrey 'a 
great mathematician', thought him something of a dull 
stick, describing him as not a pleasing companion. 30 
Godfrey calculated ephemerides for an almanac which 
Franklin printed and distributed, and he also assisted 
Logan in making astronomical observations at his home." 
Godfrey's instrument had been tried in the Truman sloopon 
voyages to Jamaica and Newfoundland in 1730-1, proving 
highly successful, whereupon Godfrey had shown the 
design to Logan, asking him to transmit it to the proper 
authorities in the hope of receiving an award under the 
Longitude Act. Unfortunately Logan took no action until he 
saw Fladley's description in the Philosophical Transactions. 
The Royal Society investigated the claims of both parties in 
January 1733 and came to the conclusion that this was 
indeed a near-simultaneous invention by Godfrey and 
1 ladley. 

But this was not the end of the story. In 1742 there was 
found among Hdmond I (alley's papers after his death a 
drawing and description of a double-reflection quadrant 
almost identical in its general arrangement with those of 
Godfrey and Hadley, designed by no less a person than Sir 
Isaac Newton. Apparently Newton had shown the design 
to Halley in 1700 but the latter foil there was no merit in it. 
Though the principle was the same as that used in a nautical 
sextant today, Newton's actual design makes it unlikely 



<;KttN\vic:n TIME 

that it would have been practicable at sea. Nevertheless, 
had it been made public earlier, a practicable nautical angle- 
measuring instrument might have been available some 
vears sooner than it actually was . 

Whoever may have been the true inventor, the instru- 
ment which was to prove the basis for all subsequent de- 
velopments for angle-measurement at sea almost up to the 
present day came to be known as Hadley's reflecting 
quadrant - or just plain 'I ladley' -and, over the next twenty 
years or so, appeared on the commercial market all over the 
world, superseding all previous instruments of that kind. 
Three subsequent developments are worth mentioning. 
First, in 1752 Tobias Mayer, an astronomer from Gcittingen, 
introduced a repeating circle, an instrument able to measure 
angles of any size and having other theoretical advantages 
as far as precision was concerned: circular instruments, 
however, never became popular among British seamen, 
being both cumbersome and expensive. Secondly, about 
1757 the English instrument-maker John Bird and navigator 
Captain John Campbell developed the sextant proper, with 
an arc of 6o° measuring angles up to 120° (instead of 
Hadley's arc of 45" measuring to 90°), with a frame of brass 
more rigid than previous wooden frames, and with tele- 
scopic sights. Finally, Jesse Ramsden's 'dividing engine' of 
1775 allowed sextants and circles to be made much smaller 
and lighter with no loss of accuracy. 

So the navigator now had an instrument with which he 
could measure lunar distances with adequate precision at 
sea. Another part of the lunar-distance problem had been 
solved, another ingredient found. 

The Nautical Almanac 

The problem of how to find longitude at sea astronomically 
- which gave rise to the need by navigators to find or keep 
Greenwich Time (or maybe Paris Time, or Cadiz Time) - 
was being solved piece by piece. Flamsteed had cleared the 
air by proving that, for practical navigational purposes, the 
Earth rotates at a constant speed. The Jupiter's satellite 
method seemed unlikely ever to be practicable at sea. The 
chronometer method, which was ultimately to prove the 
best, still awaited the invention of a timekeeper which could 


keep precise time for months on end in any climate, regard- 
less of the motion of the ship: but there were indications that 
such an invention would not be long in coming. 

For the lunar-distance method, to provide data for which 
Charles II had founded Greenwich Observatory, Flamsteed 
had produced an adequate catalogue of star positions, while 
I iadley's reflecting quadrant promised to give the necessary 
accuracy for the actual measurement of lunar distances. 
What was still needed was an adequate theory of the very 
complicated motion of the Moon so that her position against 
the background of the stars could be predicted several years 
in advance. And all these data had to be presented to the 
navigator in such a form that he could actually use them to 
find his longitude easily and without too much labour. 

This particular part of the longitude and time story - how 
the theory of the Moon's motion came to be established to a 
degree of accuracy useful to navigation - is highly technical 
and we will not attempt to describe here icliat was dis- 
covered, but will merely mention the leading characters 

It was Newton's theory of gravitation, set out in his 
Princifiia of 1687, that first offered some hope of explaining 
the irregularities of the Moon's motion. However, as Anton 
Shepherd was to sav later, '. . . but still, for want of a 
more continued and uninterrupted Series of Observations 
of the Moon, than those of Mr. Flam$hvd, the Difference of 
Sir Isaac's Theory from the Heavens would sometimes 
amount at least to five Minutes [of arcl'" - which was the 
equivalent of an error of 2'/2° in the longitude found. In the 
next fifty years or so, so important was it to 'find the longi- 
tude' that the world's best mathematicians turned their 
minds to producing the theory, and the world's best astron- 
omers, particularly those at Greenwich, to producing the 
data needed to predict the motion of the Moon. Edmond 
Halley was appointed to succeed Flamsteed as Astronomer 
Royal at Greenwich in 1720. As we have seen, Flamsteed 
had concentrated upon the stars: Halley determined to con- 
centrate upon the Moon. 

Every 18 years 11.3 days, the so-called eclipse or Saros 
cycle of 223 lunations (intervals between new moons), the 
motions of the Moon relative to those of the Sun repeat 




themselves. I {alley reasoned that the best way to predict the 
Moon's position today was to know by measurement her 
position 223 lunations ago. Therefore in 1722, at the age of 
66, he set himself the task of observing the position of the 
Moon on every possible occasion when she was visible 
crossing the meridian (that is, except for New Moon and 
clouds) throughout the 18-year cycle, thus confidently set- 
ting himself a task that could not be completed until he was 
84. He announced this in the Philosophical Transactions of 
1731. As there was then no instrument accurate enough to 
measure lunar distances at sea, he first recommended that 
the navigator should make use of that special case of the 
lunar distance when the Moon actually passes in front of, or 
very close to, the star (the technical terms for which are 
occupation and appulse), requiring only a telescope to 
observe. As a postscript to his paper he mentioned that, that 
very year, his fellow vice-president of the Royal Society, 
John 1 ladley, 'has been pleased to communicate his most 
ingenious Invention of an Instrument for taking the Angles 
with great Certainty by Reflection ... it is more than prob- 
able that the same may be applied to taking Angles at Sea 
with the desired Accuracy.'-" 

Halley died in 1742 at the age of 86. Though he had no 
regular observational help, he very nearly accomplished his 
aim in observing the Moon throughout the eighteen-year 
cycle. Alas, the accuracy of his observations - he was over 60 
when he started, remember - left much to be desired and, 
when his lunar tables were eventually published, obser- 
vations BOOn proved his predictions wrong. Meanwhile on 
the Continent the mathematicians and astronomers 
Lemonnier, Cassini de Thury, Euler, D'Alembert, and 
Qairaul were all engaged on the problem and, stimulated 
by prizes offered in 1750 and 1752 by the Academy of St. 
Petersburg, various new theories of motions and lunar 
tables were published; but all carried errors of 3 to 5 minutes 
dI arc in the place of the Moon, possibly because they were 
based on too few observations. •" 

It was a practical astronomer who finally came up with 
lunar tables of the accuracy required - Tobias Mayer of 
Gottingen, inventor of the repeating circle. Using some of 
Tiller's equations, he produced tables of the Sun and Moon 


based on his own observations and those of James Bradley 
(1693-1762), who had succeeded Halley at Greenwich in 
1742. In 1755 he sent a memorial to Admiral Lord Anson, 
Hirst Lord of the Admiralty in London, enclosing his tables. 
Thev were laid before the Board of Longitude on 6 March 
1756 with a recommendation from Bradley that they be tried 
at sea, 'but that, previous thereto, proper Instruments 
should be made to take the necessary Observations on Ship 
board, Hadley's Quadrant not being, in his opinion, 
altogether fit for that purpose'. 3S Because of the constraints 
imposed by the Seven Years War which had broken out in 
1756, the sea trials carried out by Captain Campbell in 
1757-8 were not conclusive as far as Mayer's lunar tables 
were concerned. However, the trials proved most impor- 
tant for another aspect of the lunar-distance problem, in 
that it was Qimpbell's lunar-distance observations at sea 
with a small-scale brass model of Mayer's own repeating 
circle that led to the invention of the nautical sextant we 
know today. 

It was not until 1761 that Mayer's tables were properly 
tested, by Nevfl Maskelyne (1732-1811), the future Astron- 
omer Royal, on his voyage to and from the island ot St. 
Helena where he was sent by the Royal Society to observe 
the transit of Venus. Using a Hadley quadrant (not a sex- 
tant) and Mayer's first tables, he made some very successful 
lunar-distance observations, generally achieving an 
accuracy in longitude of better than 1°. Immediately on his 
return from St. Helena, Maskelyne published his British 
Mariners' Guide, explaining in simple terms lunar-distance 
observations at sea. 1 " In fact, Maskelyne was not the first 
person to use lunar distances successfully at sea. In 1753-4 
the Abbe Nicolas- Louis de Lacaille had made such observa- 
tions on his way home to France from the Cape of Good 
Hope by way of Mauritius and Reunion, though his lunar 
tables were less precise than those of Mayer: it was his 
method of working which Maskelyne used and rec- 
ommended in his book. Also, in 1760 the Danish scholar 
Carsten Niebuhr made similar observations at Mayer's own 
request, while Alexandre-Guy Pingre followed Lacaille's 
example on his way to and from Rodriguez Island to ob- 
serve the 1761 transit of Venus." 


Tobias Mayer died in 1762 but, shortly before, he had 
prepared a new and more accurate set of tables of the Sun 
and Moon, the manuscript of which was sent to England by 
his widow and considered by the Board of Longitude on 4 
August 1763, when it was resolved that the new tables 
should be tried out by 'the Person who goes to Jamaica to 
make observations of Jupiter's Satellites'. ,8 In the event, the 
person who was sent was Maskelyne himself- to Barbados, 
not Jamaica, principally in connection with the trials of 
Harrison's fourth timekeeper. However, he did also try out 
Mayer's new tables, using one of the newly invented sex- 
tants, with which he managed to fix the position of the Isle 
of Wight to within 16 arc-minutes of the true longitude. 3 '' 

Bradley had died in 1762 and was succeeded as As- 
tronomer Royal by Nathaniel Bliss who died only two years 
later. On 19 January 1765 the Board of Longitude heard from 
the Earl of Sandwich that King George III had agreed to the 
appointment of Maskelyne himself - returned only a few 
months before from Barbados - to succeed Bliss as As- 
tronomer Royal, and he thereby became ex officio a Com- 
missioner for Longitude. At their meeting of 9 February 
Maskelyne submitted a memorial reporting his successful 
lunar-distance observations for longitude during his 
voyages to and from St. I lelena and Barbados and rec- 
ommending that the Board publish a Nautical Ephemeris to 
make the whole matter of lunar observations easy for sea- 
men: in this, he produced testimonials from four officers of 
the East India Company. 40 

Planned <md executed by Maskelyne with characteristic 
energy - he was still only 33 - The Nautical Almanac ami 
Astronomical Ephemeris for the year 1767 was published late in 
1766, and this remarkable publication has continued 
annually until the present day. in a letter to his brother 
Edmund in Calcutta, dated 15 May 1766, Nevil Maskelyne 

I he board of longitude have engaged persons to compute a very 
complete nautical & astronomical ephemeris which will come out 
next Septr. for the year 1767: and be continued annually. There will 
be 12 pages in every month. All the lunar calculations for finding 
the longitude at sea by that method will be ready performed: & 
other useful & new tables added to facilitate the whole calculation; 

OCTO'B ER' i?7 a - 

! Diftaiices of D 's Center from O , and from Stars weft of her. 








' 5 







62. 6. 55 

87. zi. o 
99. 20. 2 

HI. ?. 52 

3 Hours. 


63. 44. 49 

76. 32. 59 


100. 54. 47 

112. 34. 22 

33- °- J» 

58. 8. 6 

70. 22.45 

34- 36- n 

47. 14. 4° 

6 Hours. 


65. 22. 18 

78. 7. 10 

90. 26. 35 

102. 23. 14 

114. 6.37 

9 Hours. 

J). M. S. 

66. 59". 22 

79. 40. 56 


103. 51.23 

i.;. 26. 3* 

36.11. 47 

48. 48. 3* 
61. 12.55 

' 15. 30. 17I 17. a. 2o| 18. 34. 12 

62.45. 2 

20. 5.52 

17. Lunar-distance table, t'rom The Nautical Almanac and Astronomical 
Ephemeris h>r the year 1772. 

18. Taking .1 lunar distance. From K. Dunkin, Tlie Midnight Sky, 2nd edn. (London | iS 7 yl), 
p. 256. 


so tlmi the sailers will have link- more to do Hum to observe carefully the 
moon's distance from the sun or a proper star," which are also sel down 
in the ephemeris, in order to find their longitudes . . . 41 

It is dear from this that the most practical features in the 
new almanac were the tables of lunar distances giving, for 
every three hours throughout the year, predicted angular 
distances from the Moon's centre to selected zodiacal stars, 
or to the Sun's centre. Such tables had been suggested by 
Lacaille to cut down the amount of arithmetical work 
17 needed by navigators. And so it proved: the time taken to 

work out a sight and obtain a longitude at sea was cut from 
over four hours to thirty minutes or so. 

In 1675 King Charles II had directed his royal astronomer 
'to apply himself ... to the rectifying the tables of the 
motions of the heavens, and the places of the fixed stars 
. . .'. In 1725 Flamsteed's Historia Coelestis Britannia had 
published the places of the fixed stars. Now, ninety-one 
years after Charles's original warrant, the fifth roval as- 
tronomer N'evil Maskelyne published 'the tables of the 
motions of the heavens' in a form suitable for navigators 
"... so as to find out the so-inuch-desired longitude of 
places for the perfecting the art of navigation'. 42 So King 
Charles's directive had at last been complied with. And 
what is particularly relevant to this story of Greenwich time 
is that Maskelyne's almanac was based on the Greenwich 
meridian. Up to that time, seamen usually expressed their 
longitude as a certain number of degrees and minutes (or 
leagues) east or west of their departure point or their desti- 
nation - 3 47' west of the Lizard, for example. But now am 
navigator using Maskelyne's Nautical Almanac to find longi- 
18 tude astronomically - and a very high proportion of the 

world's deep-sea navigators began to do so from 1767 - 
must end up with an answer based on the Greenwich mer- 
idian. Indeed, from 1774 to 1788 this applied even to those 
using the official French almanac Connaissance ies Temps 
where, with Maskelyne's agreement and assistance, the 
British lunar-distance tables (based on Greenwich) were 
reprinted verbatim despite the fact that all the other tables in 
the almanac were based on Paris. 
The navigator, having obtained a Greenwich-based 

' Author's italks 


longitude, needed to plot his position on a chart. So map 
and chart publishers the world over began to provide longi- 
tude graduations based on Greenwich, to such an extent 
that, when the need eventually arose for a prime meridian 
for longitude and time to be agreed internationally, it was 
Greenwich that was chosen (rather than, say, Paris), largely 
because, bv that time, no less than 72 per cent of all the 
world's shipping tonnage was using charts based on Green- 
wich. And it was the publication in 1766 of The Nautical 
Almanac which had started the chain of events described 

The marine chronometer in Britain 

In Maskelyne's 'Explanation' to the first Nautical Almanac, 
he told why tabulations were given in apparent solar time 
rather than mean solar time. But he ended up with the 
following paragraph: 'But if Watches made upon Mr. John 
Harrison's or other equivalent Principles should be brought 
into Use at Sea, the apparent Time deduced from an Alti- 
tude of the Sun must be corrected by the Equation of Time, 
and the Mean Time found compared with that shewn by the 
Watch, the Difference will be the Longitude in Time from 
the Meridian by which the Watch was set; as near as the 
Going of the Watch can be depended upon.' 41 

The story of how John 'Longitude' Harrison (1693-1776) 
eventually received nearly the whole of the 1714 Longitude 
Act's highest award has been told often and well. 4 - 1 
Harrison, son of a country carpenter, was born near Wake- 
field in Yorkshire but moved with the family to Barton in 
I .incolnshire opposite the port of Hull at an early age. John 
began by following his father's profession but soon turned 
to clockmaking. Bv 1727 he and his brother had made two 
very superior precision clocks incorporating many new 
ideas, the most enduring of which were (a) the gridiron 
pendulum, an apparatus which remains the same effective 
length whatever the temperature (the longer the pendulum 
the slower the beat and vice versa, so for good timekeeping 
it must not change length); and (/>) the 'going ratchet', to 
keep the clock going while it is being wound up. 

He visited London about 1730 to learn more of the 
enormous awards being offered by Parliament for a solution 





to the problem of longitude at sea, one approach to which 
lay in his own held of precision timekeeping. Harrison was 
introduced to the Astronomer Royal, Halley (who was a 
Commissioner for Longitude) and to George Graham 
(1673-1751), one-time partner to Thomas Tompion and the 
most influential clockmaker of his day. Graham recognized 
Harrison's potential, lent him money, and persuaded both 
the Hast India Company and Charles Stanhope (son of the 
first Earl Stanhope, who had played a leading part in the 
passing of the Longitude Act twenty-five years before) to 
grant I Iarrison sums of money to help make his first sea 

In recent years it has been the practice to call Harrison's 
marine timekeepers Hi, H2, etc., a practice we will follow 
here. Hi, a large and heavy machine three feet high, was 
completed in 1735. Thanks to the influence of Halley and 
Graham, the Board of Longitude arranged a sea trial: 
I Iarrison accompanied Hi to Lisbon and back in HM Ships 
Centurion and Orford in 1736. Hi perfonned well but 
Harrison was not entirely satisfied and, in any case, the Act 
stipulated a voyage to the West Indies to qualify for the 
higher awards. At a meeting of the Board of Longitude on 30 
June 1737 - possibly the first ever, certainly the first for 
which minutes survive - the Commissioners granted 
I Iarrison £250 to make another 'sea-clock', with promise of 
more money if the clock was successful. H2 was completed 
in 1739 but I Iarrison was still not satisfied and proposed to 
make a third machine. Despite slow progress - H3 took 
nineteen years to complete - the Board and the Royal 
Society kept faith in Harrison, the former granting him a 
total of 1:3,000 between 1741 and 1762, the latter awarding 
him in 1749 their highest award, the Copley Medal, as 'the 
author of the most important scientific discovery or contri- 
bution to science by experiment or otherwise'. 

In July 1760, while the Seven Years War was in progress, 
1 Iarrison declared H3 ready for a trial to the West Indies 
under the conditions of the 1714 Act. But he also produced 
to the meeting a large watch, known today as H4, which, he 
said, 'answered beyond his expectation'. 45 The Board gave 
Harrison another £500, making a grand total of £3,250 since 
1737, to finish adjusting H4, so that both H3 and 1 14 could 


be tried together. In the event, Harrison decided to stake his 
all on the watch which, on i8 November 1761, was em- 
barked with his son William (John was 68) in HMS Deptford, 
bound for Jamaica. They reached Portsmouth on return in 

March 1762. 

Though I Iarrison claimed H4 had more than complied 
with the provisions of the 1714 Act, the Board thought 
otherwise, awarded him £1,500 with a promise of £1,000 
later, and said there would have to be another trial. On 8 
April 1763, following a petition from Harrison, an Act of 
Parliament received the Royal Assent promising £5,000 if he 
would disclose details of the watch to a committee of experts 
appointed by Parliament. 46 Indeed, as we shall see, the 
Academie d'es Sciences in Paris sent representatives to 
attend these disclosures, despite the fact that France was 
still technically an enemy state. (The Treaty of Paris was 
proclaimed on 22 March, ratified on 5 May 1763.) 
Maskelvne's British Mariners' Guide, advocating the rival 
lunar-distance method, was published the same year. 
Harrison saw this as a threat: someone - perhaps even 
Maskelyne himself - might win the main prize before he, 
Harrison, was ready with his alternative timekeeper 
method. He therefore decided to abandon the £5,000 
offered by Parliament and refused to disclose his secrets, 
but instead pressed for the second West Indies trial to settle 
the matter of the £20,000 reward once and for all. 

The next trial took place in I IMS Tartar, which sailed from 
Portsmouth for Barbados on 28 March 1764. with William 
Harrison and H4 embarked. Great precautions were taken 
to make the trial a fair one: Maskelyne was sent on ahead to 
settle the longitude of Barbados by observations ashore; and 
a particularly accurate clock by John Shelton (the same who 
made the clock taken to Barbados by Maskelyne) was 
borrowed at 1 larrison's suggestion from the third Duke of 
Richmond (great-grandson of Charles II and the Duchess of 
Portsmouth) for use in observations at Portsmouth before 

the voyage. * 7 

The story of Harrison's fight to get his hoped-for reward 
is too long to tell in full. Although on the face of it lU's latest 
results satisfied the 1714 Act, the Board was still not pre- 
pared to recommend the award of the full £20,000 without 

19- I larrison's first marine 
timekeeper of 1735 (Hi 
together with his prize- 
winning watch 1 14 (1759) 
ruin-), Kendall's copy (Ki, 
i7fx)) which sailed with 
Captain Cook (right), and 
Kendall's second (K2, 1771), 
taken by the mutineers in the 
Bounty to Pitcaim Island in 
1789 ftjfr). 

20. Kendall's Ki, preserved 

.it the National Maritime 
Museum. Greenwich. 


further proof that H4's results were not a matter of chance. 
In May 1765 Parliament passed another Act which effec- 
tively changed the rules of the game: Harrison would get 
£10,000 as soon as he disclosed his secrets and handed over 
all longitude machines to the Astronomer Royal - by then 
Maskelyne himself; but the second £10,000 would only be 
awarded 'when other timekeepers of the same kind shall be 
made' - and proved to be accurate enough to find the 
longitude to 30 miles. 48 The same Act authorized the 
awards to Mayer's widow and to Euler, and directed the 
Board to compile and publish the Nautical Almanac. 
Harrison was in despair. He was by then 72, his eyesight 
and general health were failing. However, he complied - 
under protest. H4 was put on trial at Greenwich and was 
then handed over to a London watchmaker, Larcum 
Kendall, to be copied, while Harrison and his son started 
making another longitude watch, Hs, which they com- 
pleted five years later. 

The storv of Kendall's copy, Ki, during Captain Cook's 
second voyage (the most important secondary aim of which 
was the testing of Ki and three chronometers by John 
Arnold (1736-99), Harrison having refused to send H5) is 
remarkable. 49 In a voyage lasting almost exactly three years, 
in the Antarctic as well as the Tropics, it performed magnifi- 
cently, the uncertainty of its daily rate never exceeding 8 
seconds of time (2 nautical miles on the equator). When the 
voyage was nearly over, Cook wrote to the Secretary of the 
Admiralty from the Cape: 'Mr. Kendall's Watch has ex- 
ceeded the expectations of its most zealous Advocate and by 
being now and then corrected by lunar observations [Cook 
had the new Nautical Almanac] has been our faithful guide 
through all the vicissitude-, of climates.'* 5 " Perhaps even more 
telling are the comments of William Wales, the Board of 
Longitude's astronomer who sailed in the Resolution with 
Cook on this same voyage: 'From the preceding account it 
appears to what an amazing degree of accuracy the in- 
genious Inventor of this watch had brought this branch of 
mechanics so long ago as the year 1762, or 3; and at the same 
time what room is yet left for future improvements by other 
Artists: but let no man boast that he has excelled him, until 
his machines have undergone as rigorous a trial as this has 

done.' 51 
* Author's italics. 




In the meantime Harrison had taken the bold step of 
approaching King George III personally. As a result, the 
King arranged for H5 to be put on trial at his private obser- 
vatory at Richmond, today called Kew Observatory. When 
the whole story was retailed to him the King is reported to 
have said: 'By God! Harrison, I will see you righted!' Finally, 
the Act 13 Geo. Ill c 77 was passed in June 1773, granting the 
80-year-old John Harrison £8,750, just £1,250 less than he 
had hoped for. (Over the years, he received a total of 
£22,550 under the Act of 1714.) 

John I larrison died on 24 March 1776, just eight months 
after Captain Cook had returned from a voyage which 
proved beyond all doubt that it was possible to make a 
satisfactory longitude watch. But those Harrison produced 
were experimental and expensive. It was left to his younger 
colleagues to design chronometers (as they started to be 
called) cheap enough to be within reach of the ordinary 
navigator. In those early days the names of John Arnold, 
Thomas Earnshaw, and Thomas Mudge were particularly 
important. They sold chronometers for 60 guineas or so 
(Mudge for rather more), whereas Kendall had received 
£500 for his copy of H4; and all received rewards from the 
Board of Longitude for their developments in the design of 
chronometers, none of which, incidentally, incorporated 
Harrison's own innovations. To Arnold must go the honour 
of devising manufacturing methods - coupled with rigid 
quality control - which might almost be called mass produc- 
tion. Arnold and his son made chronometers which were 
used in ships of all nations for fifty years or more; Earnshaw 
was not far behind. 

The British East India Company was early in insisting that 
all their ships should carry chronometers. The Royal Navy 
was somewhat slower in following suit: it was 1840 or later 
before ships carried chronometers in home waters, for 
example. But nevertheless the British Admiralty continued 
to do much to stimulate technical development, particularly 
by instituting annual chronometer trials which took place at 
Greenwich from 1821, with prizes for the best chronometers 
submitted. 52 From the very first it was to Greenwich time 
that those who used the Nautical Almanac set their chron- 



The marine chronometer in France 

But it was not only in Britain that the marine chronometer 
was being developed. In France two names stand out in this 
connection: Pierre Le Roy (1717-85), who succeeded his 
famous father as clockmaker to the King; and Ferdinand 
Berthoud (1729-1807), who was born in Switzerland but 
who spent most of his working life in France. In 1754 Le Roy 
and Berthoud each deposited with the Academie descrip- 
tions of their respective marine timekeepers. Le Roy com- 
pleted his first in 1756 but it was never properly tested. 
Berthoud's No. 1 was ready in 1763: it was tested ashore by 
the astronomer Charles-Etienne Camus when it showed 
variations up to 16 minutes in 24 hours. In 1763 also, Le Roy 
presented his second, three feet high, but this received no 
better trial than his first. 

As we have seen, the British Parliament the same year 
promised John 1 larrison £5,000 if he would disclose details 
of H4 to a committee of experts, in an Act which received the 
Royal Assent on 8 April 1763. On 21 March, four days before 
the relevant Bill had been considered by a Committee of the 
\ louse of Commons and nine days before it was passed by 
the Commons to the Lords, the Due de Nivernois, the 
French Ambassador, wrote to his superior in Paris, the Due 
de Praslin, saying that he had been told that the examina- 
tion would take place in public (which in fact was not the in- 
tention) and that he had been charged to ask 'whether you 
would wish to send a Frenchman here to be a witness and 
part of the examination. ... It is from Mr. Mackensie that 
I have all these details and his brother who is a great con- 
noisseur and protector of the arts who interests himself 
neatly in these events.' 5 ' The informant was actually James 
Stuart Mackenzie, younger brother of the 3rd Earl of Bute 
who was then Prime Minister (though he was to resign on 
18 April); both brothers were distinguished amateur 
astronomers. Nivernois's letter was passed to the Academie 
who submitted the names of Camus and Berthoud, to be 
joined by Joseph-Jerome Lefrancaisde Lalande, astronomer 
and editor of the French almanac CotmaiBsance cfes Temps, 
who happened to be already in London on a private visit. 
Lalande was a personal friend of many of those connected 
with the Harrison affair and it has been suggested that his 


presence in London at the material time had some influence 
on the apparent issuing of the invitation. S4 

Camus and Berthoud joined Lalande in London on 1 
May. Soon after, Mackenzie - who had been appointed 
Lord Privy Seal for Scotland on 15 April, just before his 
brother's resignation as Prime Minister- told Nivernois that 
he could no longer mix in the Harrison affair. On 8 May the 
French 'Commissioners' were taken to meet Harrison, 
when they were allowed to see Hi, Ha, and 1 13, but not H4. 
Lalande had seen H4 earlier, but only the outside. The 
Frenchmen waited a month in London, then grew im- 
patient. On 2 June Camus wrote to the Earl of Morton, 
vice-president of the Royal Society and one of the com- 
missioners appointed by Parliament to examine the 
I Iarrison timekeeper, saying that the French King had been 
assured that the secrets of Harrison's machine would be 
communicated to the French commissioners because the 
British Parliament wished all nations to profit from the 
discovery. In spite of these assurances, however, there was 
still no news of any plans for the disclosures to be made, 
though there was a rumour that nothing was to be done 
until Parliament reassembled. The French party, he re- 
minded Morton, had already been in London a month but, 
with school examinations looming, he personally had to 
return to Paris by mid-June and could not therefore wait for 
Parliament to reassemble. Could Morton let him, Camus, 
know when the King and the Academie should send French 
commissioners to London again to attend the disclosure of 
the secrets? 

Morton, who seems to have been acting as Chairman of 
the British commissioners, sent a reply the following day. I 
am altogether a stranger to any invitation or the repeated 
assurances which you allege were given; ... at the same 
time, if Mr. Harrison had thought proper to comply with the 
demands of the Commissioners, I for my part should have 
been extremely well pleased that you, Monsieur, together 
with Messrs De la Lande and Berthoud had favoured us 
with your presence at the Examination of his Machine. . . .' 
However, continued Morton, despite the fact that it was 
Parliament's intention that his machine should be made 
public for the common benefit of mankind, there was never- 


theless no obligation upon Harrison to divulge his secret; 
also it was a matter of doubt whether Harrison would allow 
anyone not specifically named in the Act to be a witness. 55 
The French delegation returned empty-handed to Paris 
shortly after Morton's letter; soon after that, Harrison made 
known his decision not to disclose his secrets for the time 
being. Although I.e Roy stated later that the British Admir- 
alty had invited French experts to witness Harrison's dis- 
closure in i763, s< " there is no mention of this in official 
British records so, in view of Morton's letter, it seems no 
official invitation was in fact extended, whatever the French 
Government may have believed at the time. 

In France in 1764 I.e Roy produced another marine time- 
keeper, half the size of his previous one, while Berthoud 
produced his Nos. 2 and 3, the latter being given a month's 
trial in VHirondelle corvette in October 1764 with very 
mediocre results. Early in 1766 Berthoud made another visit 
to England. Saying he had been studying English and knew 
all those in England concerned in the Harrison affair, 
Berthoud suggested to the Minister of Marine that another 
visit to learn the secrets of H4 would be worth while. He had 
been in touch with Short who had said that Flarrison would 
co-operate if offered a reward of £4,000. In the event, the 
Minister of Marine offered no more than £500, which 
1 Iarrison scornfully turned down as 'line si petite bagateW." 
But Berthoud did receive information from the watchmaker 
Thomas Mudge (1715-94), inventor of the lever escapement 
and one of those who attended the final disclosure of the 
secrets of 1 14 in August 1765. Mudge was later questioned 
by the Board about this apparent breach of security which, 
as it turned out, did not matter much as Berthoud made 
little use of Harrison's inventions in the future. 

Back in France the same year, I.e Roy presented to King 
Louis XV his masterpiece - a wonderful marine timekeeper 
of completely original design which later came to be known 
as 'A' (for anaemic). The following year, 'A' and a second 
timekeeper 'S' (for seconde) were tested at sea in a privately 
owned frigate UAurore, belonging to the Marquis de 
Courtanvaux, vice-president of the Academie Royale des 
Sciences. Results were good, but not good enough to 
qualify for the Prix Rouille offered for that year 'to deter- 


21. The Greenwich time-ball of 
1833, showing how the ball was 
hoisted to the top of the pole daily 
•it 12.58 and released by the 
assistant standing in front of the 
dock at Prom The Illustrated 
Ionian Almanack for 1845, p. 28. 

22. Le Roy's 'montre marine'. 
1766, preserved at the 
Conservatoire National des Arts 
et Metiers, Paris. 


mine the best way of measuring time at sea . . .'. 58 How- 
ever, in 1768 the frigate L'Enjouee, with the astronomer 
Jacques-Dominique Cassini on board, made a voyage of 161 
days to Newfoundland and back; 'A' and 'S' performed 
brilliantly, gaining for Le Roy a double Prix Rouille of 4,000 
livres (about £170 sterling). The Academie decided to offer a 
double prize once again in 1773 for competition on the same 
subject - the best way to measure time at sea. 

In the meantime 'Berthoud's Nos. 6 and 8 were being 
tested bv Fleurieu and Pingre in the frigate L'isis, 20, in a 
twelve-month voyage in the Atlantic. Results were me- 
diocre but the King did grant Berthoud, who seems to have 
been appointed chronometer-maker to the Navy, a pension 
of 3,000 livres (£128). It was then decided that the chrono- 
meters of l.e Roy and Berthoud ought to be tested along- 
side each other, so Verdun de la Crenne in the frigate La 
More, 32, sailed from Brest in October 1771 with the navi- 
gator Borda and astronomers Pingre and Mersains, with Le 
Roy's marine timekeepers 'A', 'S' and a watch, La petite 
route, Berthoud's No. 8, and two by Arsandeaux and Biesta 
on board. The voyage to Spain, Newfoundland, the Arctic, 
and Copenhagen'lasted a year. Le Roy's 'S' and Berthoud's 
No. 8 gave good results and the former won for Le Roy the 
1773 double prize of 4.<*x> livrcs - Berthoud declining to 
compete in view of his official position. 5 '' 

This marks the end of the experimental period for French 
chronometers. Thereafter, they began to be used regularly 
in ships at sea, particularly in voyages of exploration by 
n.i\ igators such as Laperouse, Dentrecasteaux, Baudin and 
others, who carried chronometers by Ferdinand Berthoud, 
by his nephew Louis Berthoud, and by the Fnglishman 
John Arnold. France was somewhat later than Britain in the 
manufacture of chronometers but quality was high and the 
names of Abraham Louis Breguet, his son Louis, and 1 lenri 
Motel deserve special mention. And this is how Com- 
mander R. T. Gould paid tribute to Le Roy in 1923: 

. . . the very remarkable timekeeper ['A'] . . . which stamps him 
for all time as one of the very greatest masters of horology who ever 


The exact genesis of many great inventions is hotly debated . . . 
But there can be no doubt at nil that the inventor of the modern citron- 


ometer h Kern 1 e Roy. Nothing can rob Harrison of the glory of 
having been the first man to make a satisfactory marine chron- 
ometer, one. too. which was of permanent usefulness, and which 
could be duplicated as often as necessary. But No. 4, in spite of its 
fine performance and beautiful mechanism, cannot be compared, 
for efficiency and design, with Le Roy's wonderful machine. 77m 
Frenchman, w ho was but little indebted to his predecessors, and not 
at all to his contemporaries, iiHilini, by >lnvr forte of genius, <i time- 
keeper which contains nil the essential mechanism 01 the modem 
chronometer.* 60 

The foundation of the French Bureau des Longitudes 
On 7 Messidor, year III (2=5 June 1795), people's representa- 
tive Gregoire (later, Bishop of Blois) proposed to the 
National Convention that France should follow Britain's 
example by having a Board of Longitude. In a long speech 
he quoted ThemistCK les .is saying thai whoever was master 
of the seas was also master of the world. The English had 
proved this to be so, he said, particularly in the war of 1761 . 
And because of it, she had become a great power whereas, 
by all the normal rules, she should play a merelv secondary 
role in the political order. But, said Gregoire, British 
tyranny must be stilled. And what better way than by using 
Hie methods she herself adopted? Britain realized that, 
without astronomy, there could be no commerce, no navy. 
So she had gone to incredible expense in pushing 
astronomy to the point of perfection. And it was to her 
Board ol Longitude that much of the credit for this was due: 
not only did it have enormous sums of money to disburse, 
but it also published the Xmitical Almanac - admittedly on 
a French model - which had become their seamen's 
handbook. 61 

Gregoire suggested the foundation of a Bureau des 
Longitudes, on the British model but smaller and more 
manageable, to superintend the activities of Paris Observa- 
tory and the observatory of the Ecole Militaire, and to over- 
see the publication of the Connaissanee des Temps, then, as 
now, under private proprietorship. So the Bureau was 
founded that same year, the founder members being the 
geometers Lagrange and Laplace; the astronomers Lalande, 
Cassini, Vlechain, and Delambre; the retired navigators 
Borda and Bouganville; the geographer Buache; and the 

' Author's italics 


instrument-maker Caroche, the professions of the members 
and their number having been laid down in the Law of 1795, 
which was passed immediately thanks to the eloquence of 
representative Gregoire. The French Bureau still survives 
and remains a most prestigious scientific body, super- 
intending the Paris Observatory and the Coimaissance des 
Temps, publishing an important Anmiairc, and acting as the 
French Government's chief adviser on scientific matters 
concerned with navigation. 62 

Britain's Board of Longitude was wound up in 1828, as it 
was considered that its original function, the discovery of 
the longitude, had been achieved. Nevertheless, in 1918 the 
British report on the Conference on Standard lime at Sea 
was so impressed with the achievements of the French 
Bureau that they recommended that the British Board of 
Longitude be re-established. But no action was taken. 


As the name implies, a marine timekeeper is designed to 
keep time at sea. But for navigational purposes it is necessary 
to know the time in the first place, and the going of the 
timekeeper - the chronometer - must be checked period- 
ically thereafter. In the early days of chronometers this 
could be done by lunar observations ashore or afloat (not 
very accurate), by stellar observations ashore with a sextant 
and artificial horizon, or by comparison with an observatory 
clock ashore. Whatever method was used, it was most un- 
wise actually to move the chronometer (this might disturb 
its going), so a pocket watch had to be used as an inter- 
mediary, or a signal made from ashore which could be seen 
or heard on board. In the 1820s there are several reports of 
these signals being made from shore for the benefit of ships 
in harbour - a flag dipped, a gun fired, a searchlight 
eclipsed, a rocket fired. But these all seem to have been ad 
hoc arrangements; there were no regular time signals. 

Captain Robert Wauchope, RN, seems to have been the 
first person to propose that time-balls should be erected,in a 
letter to the British Admiralty in December 1824, 63 though 
no immediate steps were taken to implement his sugges- 
tions. However, in 1833 the following Notice to Mariners 
was issued. 


Admiralty, 28 October 1833. 
The Lords Commissioners of the Admiralty hereby give notice, 
that a ball will henceforward be dropped, every day, from the top 
of a pole on the Eastern Turret of the Royal Observatory at Green- 
wich, at the moment of one o'clock P.M. mean solar time. By 
observing the first instant of its downward movement, all vessels 
in the adjacent reaches of the river as well as in most of the docks, 
will thereby have an opportunity of regulating and rating their 
21 chronometers. 

The ball will be hoisted half-way up the pole, at five minutes 
before One o'clock, as a preparatory signal, and close up at two 
minutes before One. 

By command of their Lordships 
John Barrow"' 

One o'clock was chosen for dropping the ball because, at 
noon, the astronomers might be busy finding the time. 

The apparatus, constructed in 1833 by Messrs. Maudslay 
& Field at a cost of about £180, remains substantially un- 
changed today except that, since 1852, the actual moment of 
drop has been controlled by an electric current from a 
master clock (see next chapter) and, since i960, the raising 
of the ball has also been made automatic. 65 A time-ball had 
been erected by the East India Company on the island of St. 
Helena by December 1834. Other time-balls followed. 

Not only did the Greenwich time-ball - said to be the 
world's first public time signal - give Greenwich time to 
ships in London's river and docks, but, for the first time, it 
made Greenwich time regularly available to those ashore 
who could see it, including much of London. 

4 Greenwich time for Great Britain 

Time for civil purposes 

So far we have been considering time from the points of 
view of two very limited and specialist classes of user - 
astronomers and navigators. In this chapter we shall start 
considering time as it affects people in general - time for 
civil purposes. From the earliest times man has regulated 
his activities by the Sun, as indeed have animals and plants, 
by the daily alternation of day and night, and the yearly 
alternation of winter and summer. It is the first of these 
which concerns us here - the daily alternation of light and 
darkness, the time to work and the time to rest. 

At first, the simple division of night and day sufficed for 
measuring periods of time shorter than a month (or 
'moon'). To express a period of twenty-four hours some 
civilizations used the word day (as we do in the West today), 
some the word night (as did the Jews and Arabs, from which 
usage came the F.nglish word fortnight, for example). But 
soon this primary division was found to be inadequate and 
the day began to be divided into hours, sometimes, as we 
have already seen, unequal hours based on daylight and 
darkness, sometimes equal hours which divided up the 
whole day - from sunrise to sunrise, from sunset to sunset, 
from midday to midday, or from midnight to midnight - 
into periods of equal length. 1 At a later stage it became 
necessary to divide the hour, first into simple halves and 
quarters, later still into minutes (prima niinuta) and seconds 
(secunda niinuta). 

The simplest sundials consisted of a vertical rod or pillar 
a gnomon - which sufficed to measure unequal hours. Then 
sundials proper were devised, capable of measuring equal 
hours: these were in use in Athens at least from the time of 
Pericles. But, as we have already seen, sundial time has a 


c;refnwich it me 

disadvantage which only became apparent with the de- 
velopment of clocks and watches. Because the Earth's orbit 
around the Sun is not circular, and because of the tilt of the 
Earth's axis (which causes summer and winter), the time as 
shown on a sundial - apparent solar time - is not uniform, 
varying by as much as ib minutes on either side of the mean 
during the course of the year, the amount on any particular 
10 day being known as the Equation of lime. 

This inequality of the Sun's motion in the ecliptic - that it 
seems to travel faster at some seasons than others - had 
been known since the 5th century bc when it had been 
discovered with the aid of the gnomon that the equinoxes 
and the solstices did not divide the year into four equal parts 
as might have been expected. But to the ordinary citizen this 
non-uniformity of sundial time was at first no great incon- 
venience. Until well on into the eighteenth century clocks 
and watches (which, within their limits of accuracy, showed 
»wan solar time) were mostly for the rich and, in any case, 
their inaccuracy made the discrepancy between clock and 
sundial less obvious. However, by the end of the eighteenth 
century clocks and watches became commonplace and their 
performance improved enormously. Communities began to 
keep mean time in preference to apparent time - Geneva 
from 1780, England from 1792, Berlin from t8io, Paris from 
1816. Arago recounts how, before the introduction of this 
change in Paris, the Prefect asked for guarantees from the 
Bureau des Longitudes (who had suggested the change) 
that there would be no trouble from the people at large. 
Would they not be worried that midi might not in future 
always occur in the middle of the day - which would be a 
contradiction in terms? In the event, the change occurred 
with almost no comment at all. 2 But, except for 
purposes, the time kept was still local time, based on the 
meridian of the place concerned. That this involved each 
community keeping different times - the local time in 
London and Plymouth, for example, differs by some 16 
minutes - was also no great inconvenience, because travel- 
lers were few and the rate of travel slow. 

Time distribution by hand 

The coming of the railways from 1825 onwards, however, 


brought a new situation. As far as the railways were con- 
cerned, it was extremely inconvenient, to say the least, that 
there should be no standard time throughout the system. 
And even before the railways, this inconvenience made 
itself felt in at least one field of land travel, the running of 
mail coaches strictly to timetable, first suggested in England 
by John Palmer, manager of the Bath Theatre, in 1782. The 
service started in 1784 but, by 1792, the only coaches on the 
road were those supplied by one Besant. In addition to 
curving His Majesty's Mails, the coaches were constructed 
to carry five passengers, four inside and one out. The coach- 
man was a servant of the company, the mail guard a Post 
Office employee; both wore uniform. The guard was not 
only responsible for the safety of the mails in his charge but 
had also to see that the coach kept to time. He carried 
firearms and also a timepiece, regulated, it is said, 'to gain 
about 15 minutes in 24 hours, so that, when travelling 
eastwards, it might accord with real time. Of course, in the 
opposite direction a corresponding allowance was made.' 3 
The mail coach tradition - that the person responsible for 
punctuality should carry an official timepiece - was con- 23 

tinued by the railways, though without the allowances for 
easting and westing. 

The lack of any system of uniform time was an inconven- 
ience for astronomers also. The eminent Sir John Herschel 
devised Equinoctial Time, first described in a supplement to 
the Nautical Almanac for 1828, based on his Equinoctial Year 
which, by definition, started at the moment of the vernal 
equinox (when the Sun is vertically overhead on the 
equator, on 21 March each year give or take a day). Equi- 
noctial time, a time-scale of astronomical origin but inde- 
pendent of terrestrial longitude, was thus the precursor of 
Ephemeris Time which was introduced for much the same 
reasons in 1952: the former seems seldom to have been used 
and the Nautical Almanac ceased giving data from which it 
could be obtained in 1876. 

The period from 1820 to 1850 saw far-reaching develop- 
ments in communications which were to change people's 
attitudes towards time and timekeeping: the first public 
passenger train in 1825; the first Atlantic crossing under 
steam power in 1827; Wheatstone's electric telegraph of 

8 4 


1836; mail sent by rail from 1838; Bradshaw's railway time- 
tables of 1839; Bain's electric clock of 1841; the first public 
telegraph of 1843 (running alongside the Great Western 
Railway line from Paddington to Slough). By the 1840s there 
were at least three types of organization for which it was a 
grave inconvenience, to say the least, that different com- 
munities should keep different times - the Post Office, the 
railways, and the telegraph companies. 

Luckily, as the need for a standard time became pressing, 
the means of satisfying that need became available - and, in 
Britain at least, the person to organize it was standing 
ready. The means was 'galvanism' (as electricity was then 
known), using the newly invented electric clock and electric 
telegraph; the person was George Airy (1801-92), seventh 
Astronomer Royal, one of the few scientists in England then 
in government service. But at first, means other than gal- 
vanism were used to distribute time. We have already heard 
about the Greenwich time-ball of 1833. In June 1836, the 
year the London and Greenwich Railway was opened, one 
of the assistants at Greenwich Observatory, John Henry 
Belville, started a weekly call on the principal chronometer- 
makers in London, taking with him a pocket chronometer 
set to Greenwich time. 

Belville was the son of a French widow who, after the 
guillotining of her husband in the Revolution, lied to 
England helped, it is said, by John Pond (1767-1836) who 
was travelling in France at the time and who was to succeed 
Maskelyne as Astronomer Royal in 181 1 . John Henry Bel vi Ik- 
was born in Bath in 1796 soon after Mine Belville's arrival in 
England, and when he was five years old she died. Pond 
adopted the boy and brought him up as his own son, send- 
ing him to Cambridge to complete his education and mak- 
ing him second assistant at the observatory, where he 
was known as Mr John Henry, the name Belville being 
dropped. When Airy succeeded Pond in 1835, Henry re- 
mained, dying in harness in 1856, by then Senior Assistant 
and in charge of the transit circle and chronometers. Airy 
paid tribute to him in his annual report the next year- 'the 
last who remained from Mr. Pond's establishment, and one 
of the most faithful and zealous of my coadjutors'" 1 - which, 
from Airy, was praise indeed! 

23. N'incteenlh-century train 
guards' watches. 

24. Miss Ruth Belville. 


On Henry's death the time distribution service was con- 
tinued by his widow who retired in 1892, to be succeeded by 
her daughter Ruth Belville, who continued the service until 
the 1930s. In her later years 'the Greenwich time lady' had 
some forty to fifty houses where she called once a week with 
her famous Arnold chronometer, conveying Greenwich 
time. She visited Greenwich Observatory each Mondav 

J J 

24 morning when the chronometer was checked for her and a 

certificate was issued showing its error. The same watch 
was used throughout, a large-size silver pocket chrono- 
meter, No. 485/786 by John Arnold and Son, made for one 
of George Ill's sons but rejected because it was 'like a warm- 
ing-pan'. It was subsequently selected after competition at 
the Observatory for the office it has since filled. It is now in 
the collection of the Worshipful Company of Clockmakers, 
to whom it was bequeathed by Miss Belville on her death, 
aged 90, on 7 December 1943. Donald de Carle, who talked 
personally with Miss Belville, has described her visits: 

. . . she always referred to the watch as Arnold, as if it were the 
Christian name of a dear friend. Her business with a client would 
be performed something like this: 'Good morning, Miss Belville, 
how's Arnold today?' - 'Good morning! Arnold's four seconds fast 
today' and she would take Arnold from her handbag and hand it to 
you. 'Arnold's gay (fast) today', but Miss Belville would not reply. 
The regulator or standard clock would be checked and the watch 
handed back; that would be the end of the transaction lor a week. 5 

Post Office time and railway time 

In 1840 Captain Basil Hall, RN (1788-1844), explorer and 
one-time Commissioner for Longitude, wrote to Rowland 
I lill (1795-1879), author of the penny postal system and the 
adhesive postal stamp and at that time an officer in the 
Treasury, suggesting that all post-office clocks throughout 
the different counties should be kept to London time. 

I le proposed to regulate all post-office clocks in the Kingdom, by 
means of the time brought from London by the mail-coach chron- 
ometers; and he had no doubt that, ere long, all the town clocks, 
and, eventually, all the clocks and watches of private persons, 
would fall into the same course of regulation; so that only one 
expression of time would prevail over the country, and ever)' clock 
and watch indicate by its hands the same hour and minute at the 
same moment of absolute time. 6 


So ran the report under the headline, 'Important to Railway 
Travellers. Uniformity of Clocks throughout Great Britain', 
in the first-ever number of the Illustrated London News in 
1842, a report of a talk to the Birmingham Philosophical 
Institute by Abraham Follett Osier (1808-1903), a distin- 
guished Birmingham meteorologist and businessman 
whose self-registering anemometer and pluviometer 
(measuring wind speed and direction, and the amount of 
rainfall) had been installed at Greenwich the previous year. 
Osier, demanding government action to institute 'British 
Time' throughout the kingdom, quoted Basil Hall who had, 
in turn, credited the late Dr. William Hyde Wollaston (1766- 
1828) with the original idea. 

In fact, it was not the Post Office but the railways who 
eventually forced a uniform time on a not-unwilling popu- 
lation. In November 1840 the Great Western Railway 
ordered that London time should be kept at all its stations 
and in its timetables; 7 many other railways followed suit 
during the next few years, the Midland and the South 
Eastern being examples. In 1845 the Liverpool & Man- 
chester Railway Company petitioned Parliament lo grant 
'uniformity of time for all ordinary and commercial pur- 
poses throughout the land'. 8 The petition was unsuccessful 
but in January 1846 the recently constituted North Western 
Railway introduced London time at their Manchester and 
Liverpool termini. In November the same year H. P. 
Bruyeres, their General Manager at Huston, received a 
report about the late running of a train which was attri- 
buted to the fact that London time was kept on the line 
between Rugby and York (run by the Midland Railway) 
whereas local Rugby time was kept at Rugby Station (run by 
the North Western) -and would the North Western change 
to London time, please? 9 

In 1847 Henry Booth, secretary of the Liverpool & Man- 
chester, published a broadsheet addressed to the Rt. Hon. 
Edward Strutt, Chairman of the Railway Commissioners, 
asking him to use his influence to get government action. 
According to Booth, the Post Office had already accepted 
Basil Hall's suggestions: '. . . accordingly, all their move- 
ments are regulated by "London time". By it is the great 
scheme of intercommunication adjusted, from one end of 



the kingdom to the other; the ever-varying longitude of a 
thousand post-towns is made subservient to the metropoli- 
tan chime of St. Martin's-le-Grand.' "'lie pointed out all the 
anomalies which were beginning to occur with the popula- 
tion as a whole keeping a different time from the railways 
and telegraphs - the missed trains; Bradshaw's timetable 
which, being in local time, seemed to make east-west travel 
faster than when going from west to east; the mail that left 
Holyhead at midnight on Wednesday by Holyhead time, 
which happened to be Thursday morning by London time; 
the baby born in London early on Saturday, the news of 
whose birth could be received in Dublin by telegraph on 
Friday night. Booth asked that the proposed uniformity of 
time be authorized by Act of Parliament, commencing with 
the year 1848. He added darkly: 'I am not sanguine that the 
change recommended will be without opposition from the 
Astronomer Royal, or the I Iydrographer's office.' 1 ' And so 
it proved: it was to be thirty years before there was any 
government action on the matter of legal time. 

The railways, however, had no such inhibitions. On 22 
September 1847 the Railway Clearing House (a body set up 
in 1842 to co-ordinate many aspects of railway operation in 
Great Britain) resolved 'that it be recommended to each 
Company to adopt Greenwich time at all their stations as 
soon as the Post Office permits them to do so'. 12 (The 
difference between London (St. Paul's) and Greenwich 
lime is 23 seconds.) On 1 December 1847 'he London and 
North Western and the newly-completed Caledonian Rail- 
way adopted London time 'in consequence of instructions 
received from the General Post Office'. '■' It seems likely that 
other railways in Britain conformed on the same date 
because, in Bradthaw's Railway Guide for January 1848, the 
London & South Western, London & North Western, 
Midland, Chester & Birkenhead, Lancaster & Carlisle, East 
Lancashire, and the York & North Midland Railways are all 
listed as keeping Greenwich time. We know from other 
sources that the Great Western, South Pastern, and the 
Caledonian were doing likewise. However, in December 
1848 the Chester & Holyhead earned the displeasure of the 
Illustrated Loudon /Veres which reported that the Directors had 
ordered 'that the clocks at all stations shall be regulated by 


the celebrated Craig-y-Don gun, which is 16 min. and 30 
sec. after Greenwich time'. It added: 'This cannot fail to 
prove a great inconvenience to travellers.' 14 The Irish Mail 
from London to Holyhead, however, ran to London time 
from its inception in 1848. Each morning an Admiralty 
messenger carried a watch bearing the correct London time 
which he gave to the guard of the Mail at Euston, thus 
maintaining the old mail-coach tradition. On arrival at 
I lolyhead the watch was handed to officials on the Kingston 
boat, who carried it to Dublin. On the return, the watch was 
carried back to London and handed back to the Admiralty 
messenger who met the train. IS This practice was continued 
until 1939. 

The start of the Greenwich time service 
Whatever the Astronomer Royal's private views may have 
been on whether a particular place should keep local time or 
Greenwich time, he very often stated from 1850 onwards 
that in his opinion it was a primary duty of the national 
observatory to provide Greenwich time wherever and 
whenever it was needed. In his annual report to the Board 
of Visitors for June 1849 Airy had this to say in discussing 
what changes, if any, were needed in the work of the Royal 

Another change will depend upon the use of galvanism; and, as a 
probable instance of the application of this agent, I may mention 
that, although no positive step has hitherto been taken, 1 fully 
expect in no long time to make the going of all the clocks in the 
Observatory depend on one original regulator. 

The same means will probably be employed to increase the 
general utility of the Observatory, by the extensive dissemination 
throughout the Kingdom of accurate time-signals, moved by an 
original clock at the Royal Observatory; and I have already entered 
into correspondence with the authorities of the South Eastern 
Railway (whose line of galvanic communication will shortly pass 
within nine furlongs of the Observatory,) in reference to this 
subject . . ."' 

Airy's main correspondent in the railway was Charles V. 
Walker, Telegraph Superintendent of the SE Railway Co., 
the first extant letter to whom is dated 19 May 1849" though 
there had obviously been some earlier contact. On 26 May 
Airy sent his detailed proposals. 



Things moved slowly, but 1851 brought two significant 
developments. At the Great Exhibition in London's Hyde 
Park the public clocks were driven electrically - not too 
successfully as it turned out-according to the 1849 patent of 
Charles Shepherd, of 53 Leadenhall Street. The second 
development concerned the laying of a submarine tele- 
graph cable from Dover to Calais. After an unsuccessful 
attempt in August 1850, a cable was successfully laid across 
the Channel on 25 September 1851, the news of which 
reached assembled scientists at the Great Exhibition just as 
Queen Victoria was leaving the platform after formally 
declaring the exhibition closed. 1 " The Channel cable gave 
an additional reason for Airy to pursue his plans for the 
'galvanic connection' with the SE Railway: if Greenwich 
and Paris Observatories could be directly connected by tele- 
graph, it would be possible to find the difference of longi- 
tude very accurately because observations in Paris could be 
registered on recording surfaces at Greenwich, and vice 

Briefly, Airy's plan was this: an electric clock should be 
installed at the Observatory, so fitted that it would give 
electrical impulses (11) every second for driving 'sympathetic', 
or slave, clocks (see Appendix 111) in the Observatory and 
elsewhere, and (b) every hour for dissemination of time 
signals along telegraph lines from the Observatory to 
Lewisham Station and thence along the normal railway 
telegraph lines to London Bridge Station. From there, time 
signals could be sent down the lines of the South Eastern 
Railway (SER) and also to the Central Telegraph Station of 
the Electric Telegraph Company (ETC) in Lothbury in the 
City of London for further distribution all over the country - 
to other railways, to post offices, to public clocks, and, via 
the submarine cable, to the Continent as well. Except for a 
minute or so every hour, the telegraph lines from London 
would be used for ordinary messages, but just before the 
time signal, Greenwich would be put in direct communica- 
tion with, say, Edinburgh and Plymouth, via London 
Bridge and the ETC. The clock at Greenwich, called at first 
the Normal or Motor Clock but later called the Mean Solar 
Standard Clock, had the additional facility of being capable 
of being put right electrically. The time signal itself was a 


simple electrical impulse, which could be made to ring a 
bell, drop a time-ball, fire a gun, cause a galvanometer to 
kick, operate a relay, light a light, or even put another clock 
right. And all this, by the marvels of 'galvanism', quite 
automatically, with the Greenwich clock actually pressing 
the button that made a gun fire instantaneously in, say, 
Newcastle, 280 miles away. 

In September 185 1 Airy began implementing his plans in 
earnest. Early in the month he had written to his opposite 
number at Paris Observatory; on the 19th he prepared a 
Draft of Agreement with the SER; on 6 October he started 
formal negotiations with the ETC; on 7 October he wrote to 
Shepherd asking for proposals for a suitable clock; on 26 
November he asked the Admiralty for funds (£350 or less), 

for immediately effecting a galvanic connexion between the Royal 
Observatory Greenwich and the London Bridge Railway 
Terminus, lor the throe purposes: 

1st Of regulating the principal clocks of London (the Royal 
Exchange clock and the clock shortly to be constructed for the New 
I louses of Parliament); 

2nd Of sending even- day a time signal to every part of Britain 
which is reached by a line of Galvanic telegraph; 

3rd Of communicating with the principal foreign observa- 
tories. '" 

Shepherd replied swiftly, sending a sketch and an estimate 
of £40 for the master clock and ball apparatus (for dropping 
the Greenwich time-ball automatically), with sympathetic 
clocks at £9 each. The Admiralty approved the necessary 
funds on 18 December and Airy sent a formal order on 19 
December to Shepherd for 'One automatic clock (with face 
and works) as described in your letter and drawing of 
October 18. One clock with large dial to be seen by the 
Public, near the Observatory entrance, and three smaller 
clocks: all to be moved sympathetically with the automatic- 
clock.' The total bill (of 29 September 1852) came to £244, the 
master clock itself being £70 and the wall-clock at the 
entrance £75, considerably over the original estimate. 

The telegraph lines to Lewisham Station were completed 
on 17 February 1852. On 4 June Shepherd's clock was in- 
stalled in the North Dome. On 16 July the time-ball was 
dropped electrically for the first time. On 2 August it was 



reported that the master clock was going, driving sym- 
pathetic clocks in the Chronometer Room, the Computing 
Room, the Dwelling House (Flamsteed I louse, which was 
Airy's own residence), and the Gate clock, which was set 
going on the 14th. It was also providing the impulse to drop 
25 the time-ball at 1 p.m. daily. 20 In August all was ready for a 

wider distribution of Greenwich time. The Times of 23 
August said: 

The arrangements tor transmitting true Greenwich time auto- 
matically from the Royal Observatory by electric telegraph, and 
which have already been described in The Times 1 1 1 February 1852], 
are now completed and in practical operation on the South Eastern 

At noon and at 4 p.m. a single beat or deflection of the telegraph 
needle is visible at London, Tonbridge, Ashford, Folkestone and 
Dover, which represents Greenwich mean time. The first time 
signal from Greenwich was taken experimentally by Mr. C. V. 
Walker, in the clock-room at the London terminus, at 4 p.m., 
August 5, passing down to Dover. The 1 1 a.m. signal on August 9, 
was received at London in the presence of Dr. O'Shaughnessy, of 
Calcutta, and the noon signal of the same day in the presence of 
Mr. Herbert, the secretary of the South Fastern Railway Company. 

In a letter to ALrv of 20 August Walker said that for the 
Dover line - London, Redhill, Tonbridge, Ashford, Folke- 
stone, and Dover - time signals would be taken at noon and 
4 p.m. daily. For the North Kent line - London, Lewisham, 
Blackheath, Woolwich, Frith, Gravesend, and Strood - 
signals would be taken at 2 p.m. The other twenty-one 
hourly signals would be available for the ETC at Lothbury 
for their own time-ball in the Strand (see below) and for 
distribution throughout the kingdom along the various 
other railway lines. In addition to the hourly signal, 
Shepherd's clock at Greenwich sent impulses every second 
to drive a sympathetic clock at London Bridge which auto- 
matically made the various switching operations needed for 
sending out the hourly time signals. 

Procedure at the Royal Observatory 

In essence, Airy's time distribution system at the Observa- 
tory itself was this: 


(1) Find the time bv astronomical observations of the so- 

25. Greenwich Observatory 
about 1870, showing the time- 
ball and electrically-driven 
24-hour gate clock, both 
controlled by Shepherd's 
'motor clock'. The time of the 
photograph is a few minutes 
before 19.00 astronomical 
time, which was 7 a.m. civil 

26. Shepherd's mean solar 
I standard clock and apparatus 
for dropping the lime-ball. 
From The Graphic, 8 Aug. 

94 C.KI I \WI(H riME 

c.illed 'clock stars', using the transit circle (sec Appendix II), 
generally at night - every night - when weather permits. 
26 (2) Correct the standard clock to show the time just found (or 

estimated if observations were clouded out), daily, immedi- 
ately before the principal time signals at 10 a.m. (1 p.m. 

(3) Send out the time signal, every hour, on the hour. 

(4) Repeal the procedure the following day. 

Though new clocks were brought into use and radio came 
to augment the telegraph as a means of distribution, Airy's 
1852 system, based upon the idea that the rotating Earth is 
the fundamental time-keeper, remained virtually un- 
changed until the advent of the atomic clock in the 1960s. 

Nationwide distribution of time signals 

At the start of the time service, all time signals went through 
the London Bridge switch-room and were sent from there 
(at the hours not needed by the SER itself) to the ETCs 
Central Telegraph Station at I othbury in the City of London 
for distribution to other railways, post offices, etc. When the 
underground cables from the Observatory to Lewislnim 
became defective in 1859, overhead lines were installed 
direct to Lothbury (and later to the London District Tele- 
graph Co.'s headquarters as well) so that the telegraph 
companies' signals no longer had to go via the SER. But a 
line also went via Greenwich Station to the SLR for the Deal 
time-ball (see below) at 1 p.m., for the British llorological 
Institute (for chronometer makers) at 2 p.m. and 8 p.m., and 
for the SLR itself at all other hours, on the hour." From 
London onwards, the time signals went to remote stations 
on telegraph lines used for normal messages and it was 
necessary to clear the lines of other traffic for a few minutes 
either side of the time signals (the principal ones being 10 
a.m. and 1 p.m.), so that the signal from Shepherd's clock at 
Greenwich could pass direct to, say, Glasgow at precisely 

At first, the necessary switching was done at Lothbury by 
a clock similar to that at London Bridge, though not con- 
trolled from Greenwich; but, about 1864, ^n apparatus to 
perform this switching automatically was designed by 
Cromwell F. Varlev, electrician to the Electric and Inter- 


• — 

5n i' FRENCH'S 
& 1 a" 

27. The time-ball in Cornhill, 
London, i860. From .1 letterhead of 


28. Single-needle telegraph instrument, common]) used 
in the nineteenth century for sending and receiving time 

|. French (late Bennett), chronometer signals. PramThelttusmucdi •'»•'"•• *'••■■*■ # *»"'■ ,H 74- P 

maker, in RCO MSS. 1 isi/i45 v - 5°4- 

29. The Chronopher Ro 
the Central Telegraph Office, 
General Post Office, St 

\l.1rlins-k-Cr.1nd. 187.1. The 
old chronopher is behind the the new one to hi-- leil 
Prom The Illustrated London 
\,;r<, W Dee. 1874. p. 576. 


national Telegraph Company (E & ITC: the ETC and the 
International Telegraph Company had merged in 1855). 
Driven by an accurate pendulum clock which could be kept 
correct mechanically without having to touch the pendu- 
lum, the 'chronopher' (from xpovos, time; and <t>epu>, I bear) 
placed the telegraph lines concerned in direct touch with 
Greenwich 1 minute 50 seconds before the time signal and 
put Greenwich out-of-circuil 1 minute 20 seconds after the 
hour. The original chronopher had two circuits, one serving 
stations in London, one provincial stations. In 1864 the local 
circuit sent hourly time signals to a clock and six bells within 
the Central Telegraph Office building; to time-balls at the E 

27 & ITC's office in the Strand and Bennett's of Comhill (the 
City Observatory); to clocks at the General Post Office, 
Lombard Street post office, and Dent's, the clock and 
chronometer maker in the Strand; and a signal to the 
Westminster clock ('Big Ben'). The provincial circuit sent 

28 daily time signals to telegraph offices in Manchester, Liver- 
pool, Birmingham, Glasgow, Bristol, Portsmouth, Bath, 
Cardiff, Brighton, I lull, Derby, and Lowestoft, to the royal 
residence at Sandringham, and to the London, Chatham 
and Dover Railway.-' When the Central Telegraph Station 

29 moved to St. Martin's-le-Grand in 1874 after the telegraphs 
were taken over by the General Post Office, a second and 
larger chronopher was added to the system. 

The ETC's Strand time-ball 

In 1852 the ETC's West End office was in 448 West Strand, a 
building (of which the facade survives) designed by John 
Nash with two 'pepper-pot' towers at each end, immedi- 
ately opposite today's Charing Cross Station. Edwin Clark, 
ETC's chief engineer, wrote to Airy on 27 Eebruary explain- 
ing his plans for a time-ball on one of the 'pepper-pots', to 
be dropped at 1 p.m. daily by an impulse from the Green- 
wich clock. As The Times explained while it was being 
erected in June, there was to be a time-ball 'which is in- 
tended by means of sympathetic electrical action to fall 
every day simultaneously with the well-known ball on the 
top of the Greenwich Observatory, between which and the 
Strand the electric wires have been completed for the 
purpose, so as to indicate to all London and the vessels 

.Station. 24 

A similar order of the same date is known to have been 
passed to all stations of the Great Western Railway 27 and 
doubtless, through the ETC, to other railways as well. 



below bridge exact Greenwich time.' 24 The ball was 5 feet in 

diameter, made of zinc, painted black with a broad white 

band around it, the shaft being topped with a weather vane 

with 'ETC on it. The whole cost £1,000. The automatic 

dropping of the Strand time-ball started on 28 August. 'The 

public assemble in crowds and the chronometer-makers 

think it a great boon.'" However, there were troubles. On 

29 August, with great crowds watching, it dropped 28 

seconds late; on 1 November it dropped iVi minutes early. 

Never a great success because of technical difficulties, it was 

discontinued after a few years. 

On the railways, however, all seems to have gone well. 

On 30 October the following general order was passed to 

the SER: 

Electnc Telegraph, 

Tollbridge, October 30th, 1852. 

South Eastern Railway 

General Order 


The Astronomer Royal has erected Shepherd's Electro-Magnetic 
Clock at the Royal Observatory, for the transmission of Greenwich 
Mean Time to distant places. 

On and after November i*t, the needle of your Instrument Will 
move to make the letter N precisely .it .o'clock every da v. 
[Different stations received time-signals at different hours.] 

Abstain from using the instrument for Two Minutes before that 


Watch the arrival of the signal; and make a memorandum, tor 
your own information, of the error of your Office Clock. 
' You are at liberty to allow local Clock and Watch Makers to have 
Greenwich time, providing such liberty shall not interfere with the 
Company's service and the essential privacy of telegraph Offk es, 
and the business connected therewith. 

Engineer and Superintendent 
of telegraphs 

To Mr 

K>. i hi- time-ball In the Strand, London. 
From The Illustrated London News, 1 1 
Sept. 1852, p. 205. 


The Deal time-ball 

Writing to Airy on 12J.1nu.1ry 1852 C. V. Walker said: 'Has it 
ever occurred to you to cause the Greenwich clock to drop a 
ball at Dover? It would not be impossible: it would be 
useful.' 28 On 13 April Commander Thomas Baldock wrote 
to the Admiralty suggesting that a time-ball should be 
dropped by galvanic current on one of the South Foreland 
lighthouses so that ships in the Downs, where those out- 
ward-bound waited at anchor for a favourable wind, could 
check the errors of their chronometers before taking de- 
parture. 29 

In the event, it was decided that the ball should be erected 
in the old Navy Yard at Deal. Airy ordered a galvanic clock 
and ball apparatus from Shepherd on 12 November 18^3 - 
£21 as per estimate. This time-ball came into operation on 1 
Januarv 1855, dropped by the 1 p.m. current from Green- 
wich, and continued until 1927. A return signal was sent 
from Deal to Greenwich giving an indication whether or not 
the ball had dropped on time. Soon afterwards, a time-gun 
was installed at Dover Castle, fired at noon daily by a 
current from Greenwich. 

Airy recalled the institution of the system in a properly 
patriotic address to the British Horological Institute in 1865: 

I can hardly sav how the time signal system came to be first 
proposed, it was somehow, partly in conversation, partly in other 
wavs, how, I cannot exactly say, but to Mr. C. V. Walker, Mr. Edwin 
Clark, Mr. Latimer Clark, and afterwards Mr. C. F. Varley, is the 
existence of the system mainly due. 

The Deal time ball was not proposed by me, though I have taken 
great pains to render it efficient. I have indeed always considered it 
a very proper duty of the National Observatory to promote by 
utilitarian aid the dissemination of a knowledge of accurate time 
which is now really a matter of very great importance. 

The practical result of the system will be knowledged by all 
those who have travelled abroad. We can, on an English railway, 
always obtain correct time, but not so on a French or German 
railway, where the clocks are often found considerably in error. 

Early developments in the time-distribution system 
Developments in the first twenty years of the Greenwich 
Time Service, too numerous for each to be treated in detail. 



100 GREENWICH 1 1 Ml 

can be summarized as follows: 30 

7852 Airy's chronograph installed at Greenwich, auto- 
matically recording seconds impulses (later, two-seconds 
impulses) from the Sidereal Standard Clock on the same 
paper chart as the precise times of astronomical observa- 
tions on transit circle. It did not come into full operation 
until 1854. 

7855 Four clocks in London post offices were regulated by 
signals from Greenwich, sending return signals to ensure 
correct functioning. Writing in 1868, Airy said: 'In the 
clocks at the Lombard Street Post-Office, I some years ago 
arranged an apparatus by which at noon everv day a 
galvanic current from this observatory seizes the second- 
hand of the principal clock and turns it round (if necess- 
ary) so as to make it point to oS. This apparatus has been 
long in action without a failure.'- 11 

7855 (]uh/) The Electric Telegraph Co. (1846) and the 
International Telegraph Co. (1852) merged to form the 
Electric and International Telegraph Co. (E & ITC). 

1856-7 Time-balls controlled by signals from Greenwich 
erected at the E & ITC's office in Liverpool and at the 
premises of French (late Bennett), watch and clock- 
makers, in Comhill in the City of London (with the rather 
grandiose title of the City Observatory); the same current 
dropped time-balls at Greenwich, the Strand, Cornhill, 
Deal, and Liverpool. 

7859 Underground telegraph lines from Observatory to 
Lewisham failed. Replaced by six overhead lines to 
Greenwich Station. 

1860 Time superintendent's desk set up at Royal Obser- 
vatory with instrumentation permitting all operations 
connected with time distribution to be controlled from 
one place. 

I860 E & ITC's Central Telegraph Station moved from 
Lothbury to Telegraph Street. 

7867 Time-gun controlled by Royal Observatory, Edin- 
burgh, set up at Edinburgh Castle. Time-ball on Calton 
Hill already in operation. Airy's own list of time-balls 
extant in 1861 is reprinted in Appendix V. 

7862 Great clock at Westminster - popularly 'Big Ben' - 
installed. Greenwich sent hourly time signal, with return 


signal twice daily. The clock was in no way controlled 
from Greenwich. 

7862 London District Telegraph Co. (LDTC; founded 1859) 
received Greenwich time signals for distribution in 
London, particularly to chronometer makers and 

7863 Time-guns established in old Norman keep at 
Newcastle, and at North Shields, fired at 1 p.m. by 
current from Greenwich via E & ITC (later via GPO). 

c.7864 Varley's first chronopher installed in the new 
Central Telegraph Office in Telegraph Street in the City of 

By 7865 Clock installed at factory of Warren de la Rue in 
' Bunhill Row, London, who '. . . estimates the annual 
saving to his firm by having exact time and enforcing 
strict attendance on his work-people, at £300 (besides 
some saving of gas and coals not taken into account) 
which is an amount that would otherwise be entirely lost, 
and of this he is able to make a return to his work-people 
in the way of additional privileges as respects holidays'. 32 
7865 (28 Jan.) and 7866 (77 Jan.) Snow and gales brought 
down all overhead lines from Observatory - two years 


7866-7 Underground telegraph lines laid from Observa- 
tory to Greenwich Station. 

7870 (1 Ian.) All electric telegraph companies taken over by 
General Post Office, including time signals distributed by 
E & ITC and LDTC. 

7872 (29 luly) Post Office circular confirming that 
Greenwich time was to be kept in all post offices. 

7874 (January) GPO's Telegraph Department moved from E 
& ITC's Telegraph Street office to a new building facing 
the GPO at St. Martin's-le-Grand. New and more elabor- 
ate chronopher for sending time signals in sixty different 
directions installed for 10 a.m. signal, old chronopher 
being used for the 1 p.m. signal. The following account 
by Mr. H. Eaton of Post Office Telegraphs shows time- 
distribution state in 1874: 

The Greenwich current is received hourly- This hourly current 
is transmitted to 10 subscribers (mostly chronometer-makers) in 
London. The method of observing the current varies, and is 


fixed by the subscriber, in two cases, time-balls are dropped on 
the top of the buildings; in some other cases, model time-balls 
.ire placed in the windows; and others again use an electric bell; 
while two or three have a simple galvanometer, and observe 
from the deflexion of the needle. 

The Westminster clock records its correctness and errors at 
Greenwich, as does also the clock at the Lombard Street Post 

The 10 a.m. current is most extensively used for the 
Provinces. It is transmitted automatically to 21 provincial towns 
in England (where there are subscribers), to Guernsey, 
Edinburgh, Glasgow, Dublin and Belfast. In addition to the 
automatic sender, a sound-signal is established in the Instru- 
ment Room here; when heard, a current is sent by the clerks to 
over 600 offices in direct communication with the Central 
Telegraph Office, including the principal railway termini. Many 
of these offices re-distribute the time-signal to the offices radi- 
ating from them, so that practically from the 10 a.m. current 
from Greenwich most of the post-office and railway clocks in 
the Kingdom are regulated. 

The 1 p.m. current is transmitted automatically to nine 
provincial towns, viz., Newcastle, Sunderland, Middlesboro'. 
Kendal, Hull, Norwich, Stockton. Worcester, and Nottingham. 
At the first four named, guns are fired; at the others, the current 
is observed by means of time-balls or gal variometers. 

With regard to the 10 a.m. current, I should have said that 
there is no rule as to the method of observing; the subscribers 
use the form of apparatus most suitable to themselves. At the 
Telegraph Office the signal is recorded or observed on the 
telegraph instrument." 

1888 Astronomer Royal proposed to discontinue 10 a.m. 
time-signal. Post Office objected: 'We cannot dispense 
with the 10 o'clock current from Greenwich. . . . Every 
office throughout the United Kingdom is supplied with 
this time current by a system which has taken many years 
to establish. . . . But for what purpose was Greenwich 
Observatory established, if it was not for the production 
of accurate time for national and imperial objects; and 
what object is of more consequence to the Government 
than the distribution of accurate time throughout the 
three Kingdoms to every post office and railway station? 
. . .' 34 The Astronomer Royal relented and the 10 o'clock 
signal continued. 

32 Small time-ball, about iSss, and galvanometer, about "9"<>. examples of 
tparatus often set up in jewellers' windows so that the public could get me 

apparatus often set up 
Greenwich time signal. 

33, A hatter's time signal at George 
Cirterfc Sons Ltd.. 211-17OW Kent 
Road, London, S.E. From the Soulh-East 
London and Kentish Mercury, 27 Feb. 1975- 

34. Sir George B. Airy. KCB, IKS. From Punch, 
'1883. p. 214. 


Time signals to private subscribers 

From the early days of the Greenwich Time Service the 
telegraph companies were prepared to rent a private wire 
specifically for receiving the Greenwich time signal, a 
service particularly appreciated by public bodies and 
chronometer-makers in London and the provinces. At the 
time of the transfer of the telegraphs to the State (1870) the 
Post Office took over the contracts then in force, consider- 
ably adding to their number thereafter. The charges were 
first published in the Post Office Guide of 1 January 1873, 
varying according to the subscriber's distance from the 
main post office. In London within two miles of the GPO at 
St. Martin's-le-Grand, for example, subscribers could 
receive an hourly signal for £15 a year. Provincial sub- 
scribers '/a mile from their local Head Post Office paid £12 
annually for the io a.m. current, £27 for the 1 p.m. current; if 
a mile away, the charges were £17 and £32 respectively. The 
GPO continued to offer this service until 1927. 

Subscribers had to provide their own terminal equip- 
ment -for example time-balls on buildings, small time-balls 
31, 32 in shop windows, bells, galvanometers, and so on. One of 

the more bizarre time signals was that of the hatter George 
Carter of 21 1-17 Old Kent Road, London SE, which, from 
about 1900, consisted of an elephantine silk hat outside the 
shop which, after climbing slowly upa mast, fell at precisely 
33 1 p.m. on receipt of the Greenwich time signal. In response 

to changes of fashion, the top hat was replaced after the 
First World War by a figure with a bowler hat which was 
raised and allowed to fall at one o'clock daily.- 15 

In 1876 the Standard lime Company was formed by the 
chronometer-makers Barraud & Lunds of Cornhill, 
London, principally to exploit a method of synchronizing 
clocks invented by J. A. Lund. The company received the 
hourly time signal from Greenwich by direct private wire 
and redistributed it to subscribers by its own overhead 
wires in the London area. Though the signal was principally 
designed to synchronize subscribers' master clocks (which 
could then be used to drive slave clocks) by the forcible 
correction of the minute hands every hour, it could also be 
used to operate any form of time signal, audible or visual. 
Though the bombing of London in 1941 forced the company 


to abandon overhead wires, the system continued to 
operate until i964.- ,b Somewhat similar systems operated in 
Edinburgh, Glasgow, and Liverpool, initially using time 
signals from local observatories. These British systems were 
reported to be far superior to the pneumatic system started 
in Paris about the same time, where the clocks furthest from 
the depot were said to be appreciably slower than the closer 
ones, owing to the slow build-up of pressure in the tubes." 

Railway time v. local time 

But meanwhile the matter of a uniform time - of railway 
time v. local time - had been under debate in the country as 
a whole. Soon after A. F. Osier had given his lecture in 1842, 
he proposed establishing a standard clock for Birmingham, 
and he collected funds with which he procured one of the 
highest class, made by Dent, which was placed in front of 
the Philosophical Institution. By-and-by, when he had fully 
established the clock's accuracy in the public's estimation, 
on one Sunday morning he altered the clock from Birming- 
ham time to Greenwich time without mentioning it to any- 
one, and, though the difference was remarked upon, the 
Church and private clocks and watches throughout the 
town were one by one adjusted to Greenwich mean time, 
though at that period the country generally was still keeping 
local time. 38 

The first excursion train ran in 1844 and the great increase 
in railway mileage in the late 1840s revolutionized the social 
life and habits of the country. In a leader on 12 January 1850 
The Times said:'. . . there must be thousands of our readers, 
we are sure, who, in the last three years of their lives, have 
travelled more and seen more than in all their previous life 
taken together. Thirty years ago not one countryman in 100 
had seen the metropolis. There is now scarcely one in the 
same number who has not spent his day there. Londoners 
go in swarms to Paris for half the sum, and in one-third of 
the time, which in the last reign it would have cost them to 
go to Liverpool. . . .'- 19 

But the event which caused the greatest increase in 
passenger traffic on the railways was the Great Exhibition of 
1851 in Hyde Park, London, which resulted in travel in 
Britain on an unprecedented scale, The Times reporting over 


six million visitors (only 75,000 were foreigners) - and 
almost all of those travelled by train. 4 " As we have seen, 
almost all the major railways had adopted Greenwich time 
by 1847 and, like Birmingham, many cities in the North and 
Midlands began to set their clocks to 'railway time'. In 
Scotland, in letters dated 30 November 1847, me Chairman 
of the Edinburgh and Glasgow Railway Company wrote as 
follows to the Lords Provost of the two cities: 

I am requested to bring under your notice and that of the Council 
over which you preside, the intention announced by the principal 
English Railway Companies to adopt Greenwich instead of local 
time on their lines, to which alteration the Corporations of Liver- 
pool and Manchester have resolved those towns shall conform. 

The directors of the Edinburgh and Glasgow Railway believe 
that it would be very advantageous to adopt the same time and will 
gladly do so if the authorities of the two cities which their lines 
connect agree to this change. 41 

The respective town councils agreed almost without demur, 
prompted by a leading article in The Scotsman of 4 December 
supporting the idea and forecasting that such a change 
'would not disturb the arrangements of business and 
domestic life more than the errors of our clocks do now'. 
The change took place on the night of Saturday 29 January 
1848 (the date chosen by the Post Office authorities), 
Edinburgh, Glasgow, Greenock, Stirling, and Perth all 
resetting their public clocks on the same date. 42 The change, 
which meant that clocks were put forward 12 1 1 minutes in 
Edinburgh and 17 minutes in Glasgow, seems to have met 
with virtually no public opposition, the average citizen 
being quite content to regulate his life by whatever time the 
public clocks - or the factory hooter - gave him. Of course 
there were some letters to newspapers deploring the 
change - any change. For example, Blackwood's Magazine of 
March 1848 carried an article entitled 'Greenwich Time: 
"The time is out of joint - oh, cursed spite" {Hamlet)'. In 
eight long pages the anonymous author castigated the 
Edinburgh Town Council for usurping the power of the 
Almighty. 'What in the name of whitebait have we to do 
with Greenwich more than with Timbuctoo, or Moscow, or 
Boston, or Astracan, or the capital of the Cannibal Islands? 
The great orb of day no doubt surveys all these places in 


turn, but he does not do so at the same moment, or minute, 
or hour. . . .' The article ends with a sentiment which has 
an almost modern ring about it: 'It would much conduce to 
the com fort of the lieges, if, instead of directing the course of 
the sun, you were to give occasional orders for a partial 
sweeping of the streets.' 43 Predictably, there were as many 
letters approving of the change, pointing out that the local 
time previously kept was not in fact 'God's time' but a time 
based on a fictitious Mean Sun, chosen because the Sun 
itself was such a bad timekeeper: indeed, in some months of 
the year, the new railway time was closer to God's time than 
belore. 44 

However, in the east and west of England, opposition 
continued. On 21 June 1851 there appeared in Chambers 
Edinburgh journal (paradoxically published in London) .1 
light-hearted article entitled 'Railway-time aggression'. The 
anonymous author started by saying: 'There is an aggres- 
sion far more insidious in its advances than the papal one 
[C£/ was a great supporter of the Established Church|. . . . 
We are now, in many of our British towns and villages, to 
bend before the will of a vapour, and to hasten on its pace in 
obedience to the laws of a railway company! . . ' 45 1 le then 
went on to recount some of the things that were happening, 
not specifically because railway time was being introduced 
but because there were two different kinds of time being 
used simultaneously - the bride who arrived at the Church 
at railway time while her groom (and the pastor, organist, 
and choir) arrived at local time; the ruination of a dinner 
party because of the different times kept by host and guests; 
the friction ol town (railway time) and country (local time); 
the difference in church clocks between High Church (local) 
and Nonconformist (railway). 

On 2 October 1851 'E.S.H.' started a spirited correspon- 
dence in The Tones: 

I. In trie telegraph and local lime 

Sir, - Contemporaneously with the advance of railroads, and the 
invention of the electric telegraph, the difference of time arising 
from the variation in longitude of places has been considered 
objectionable; and, lor convenience' sake, an uniformity of time - 
that of Greenwich has been adopted throughout the Kingdom, 
with exception of a few places in the west of England. 



By reason of the submarine telegraph, Kngland will now be 
brought into immediate communication with France and the 
greater portion of Europe. The question therefore arises, what 
meridian should be determined on for universal adoption? 46 

Swiftly, on 7 October, there came a riposte from 'Chronos' 
of Greenwich, thought by many to be the Astronomer Royal 

load time 

Sir, - Your correspondent 'F..S.II.', after remarking that 'a uni- 
formity of time - that of Greenwich has been adopted throughout 
the Kingdom, with exception of a few places in the West of 
lingland, the difference of lime arising from the variation of longi- 
tude of places having been considered objectionable', anticipates 
from the submarine telegraph to France an extension of this 
practice, and proposes the question, 'What meridian shall be 
determined on for universal adoption?' 

Surely, Sir, we may rather, on the contrary, hope that the facili- 
ties thus afforded for its exaggeration will make the unreasonable- 
ness of the custom referred to more glaring, and will lead to a 
return to the only true and simple rule of keeping our clocks right 
instead of keeping them wrong, as we all now do, with the 
exception of a few wise men in the west. The absurdity of calling it 
noon at a quarter to 12 is not strikingly obvious; but every one must 
see the absurdity of calling it noon at sunrise in one place, at sunset 
in another, at midnight in a third. 

It so happens that the world takes 24 hours to revolve on its axis. 
The fact may be 'considered objectionable'; but so long as it 
remains unaltered it is simply impossible that it should be the same 
hour in two different places at once; and 'uniformity of time' is as 
impracticable as uniformity of locality. 

Truth and common sense cannot long be violated with impunity; 
and great inconvenience and confusion will inevitably arise (and 
shortly, too, with the aid of the submarine telegraph), unless we 
return to the plain fashion of calling it noon when it is noon, and 
not when it is not, and accept the variation in longitude in places as 
an irresistible fact, instead of voting it 'objectionable', and pretend- 
ing to ignore it. 

The convenience of central regulation need not, of course, be lost 
by keeping true local time. A timekeeper 10 minutes west of 
London, on receiving his electric intimation that it is noon at 
Greenwich, sets his clock at 10 minutes to 12, which is just as easy 
as setting it to 12. The only persons who derive any real benefit 
from the so-called 'uniformity' are the clerks who settle the railway 
time-tables, who are thereby saved some 20 minutes' trouble in 


calculating the allowance of time to be made as trains proceed east 
or west. This appears hardly a sufficient ground for bewildering all 
the timepieces and headpieces in Kurope; but these gentlemen can 
do with us what they please, and we must be content with humbly 
entreating what we dare not demand, that they will cease from 
their desperate attempts to 'annihilate both Space and Time' 
(which have not even the laudable effect of 'making two lovers 
happy' ), and allow the old legitimate King lime to resume his place 
in our clocks and bosoms. - " 

Which only goes to prove that The Times has long been able 
to embrace many opinions, even when the dispute is over a 
subject embodied in its own name. 

In January 1852 the Illustrated London News announced 
Airy's plans for time signals by electric telegraph, noting 
that the only towns of consequence still holding out against 
Greenwich time were mainly in the east and the west of the 
kingdom - Norwich, Yarmouth, Cambridge, Ipswich, 
Colchester, Harwich; and Oxford, Bristol, Bath, Ports- 
mouth, Exeter, Dorchester, Launceston, Falmouth. 48 In 
Oxford, according to B. L. Vulliamy, clockmaker to the 
Queen, the inconvenience of two different times was met 
on the great clock on Tom Tower at Christ Church by em- 
ploying two minute hands, one set to Oxford time, one to 
Greenwich. 49 

A place where the debate waxed strongly was Exeter. On 
13 November 1851 H.S.E. of Exeter - almost certainly the 
publisher of the map - wrote to The Times to say that, at a 
recent meeting, the Town Council had resolved by 16 votes 
to 5 that 'public clocks in this city should show and strike 
Greenwich instead of local mean time'. But the Dean and 
Chapter had refused to alter the cathedral clock on which 
all other public clocks relied, one objection being that 
Plymouth had altered theirs and then altered back. (In fact, 
this was because the current Plymouth tide tables were 
based on local time but, as was pointed out, this could very 
easily be changed in the future.) H.S.E. demanded legisla- 
tion on the matter of legal time. 50 

In August 1852 the electric telegraph was about to reach 
Exeter -and, indeed, Plymouth also -and the local authori- 
ties decided to raise the subject of time once again. Sir 
Stafford Northcote (1818-87), a prominent local resident in 




m & A -^Mf 



35. Towns (in italic) in Great Britain still keeping local time in February 1852. The scale along 
the top shows the number of minutes' difference between local and Greenwich time. From 
the edition of a map published bv Henrv Ellis & Son, Exeter, in February 1852 (RGO MSS. 

hiteCt of Tom Tower was Christopher Wren, who compared it with the Great Room 
nwich (p. 36). Photo in volume of prints of Christ Church in the Bodleian Library, 

The aid 

at Greenwich (p. 36) 

Oxford (G.A. Oxon. a. 50, no. 204. p. 1 12). 


favour of the change who rose to become Chancellor of the 
Exchequer in Disraeli's Government of 1874-80, wrote to 
the Astronomer Royal in August 1852 to ask his views on 
the whole question - and specifically to ask whether he was 
the Chronos who wrote the letter of 7 October 1851 quoted 
above. Sl No! said Airy, he was not Chronos. 'My opinion on 
the question is not very distinctly formed.' I le then gave 
about three lines in favour of the change, followed by seven 
pages giving the opposite view, including the words: 'If I 
had the power of legislating in a comprehensive way, 1 
should certainly preserve local time in every place'"" 2 - 
which today seems a curious attitude in view of his sub- 
sequent declaration that the dissemination of accurate time 
was one of the prime functions of the national observatory, 
and his boast only nine months later: 'I cannot but feel a 
satisfaction in thinking that the Royal Observatory is thus 
quietly contributing to the punctuality of business through 
a large portion of this busy country.' 53 

Immediately after this exchange, the Western Luminary of 
31 August 1852 reported that another Council meeting on 
the subject had been held. Towns in the West Country 
which did keep Greenwich time - Taunton, Tiverton, 
Torquay, for example - felt no inconvenience; in other 
towns in the kingdom where Greenwich time had been 
adopted the change had been forgotten only a few days 
after the alteration had been made; for countrymen, the 
change from apparent (sundial) time to mean (clock) time 
had had an even greater impact than the change from Local 
Mean Time to Greenwich Mean Time. 54 On 28 October a 
public meeting was held at the Guildhall, Exeter, where it 
was unanimously resolved that public clocks should be 
altered. The Dean gave way: the cathedral clock was 
advanced 14 minutes to Greenwich time on 2 November 
1852, the day after the first regular daily time-signal was 
sent down the railway telegraph lines direct from Green- 
wich. 55 

In Bristol, also, there was considerable debate, turning 
partly upon the business of tide tables. The Bristol rimes of 
20 March 1852 reported that no serious inconveniences had 
arisen at Newport, Swansea, Southampton, or Liverpool. 
Bristol therefore made their decision to regulate clocks to 


Greenwich time more than a month before Exeter, at a 
meeting of the Council on 14 September 1852, though three 
inveterate admirers of ancient ways protested against the 
innovation. It is said that one of the more determined anti- 
quarian councillors continued to regulate his affairs by local 
time for many years. 56 

In Plymouth a meeting of the Town Council took place on 
the subject the following day. Plymouth, unlike Bristol, 
adopted no resolution but some of the debate reported in 
the local paper had a surprisingly modern ring to it: 

Mr. W. Moore disapproved of the movement, for workmen would 
thereby be enabled to leave work 16 minutes earlier every dav, but 
he was sure employers would not be able to get them to come 16 
minutes earlier in the morning (11 laugh). 

Mr. R. Rundle so far differed from the last speaker, that he 
believed tradesmen and others would be very glad of .1 change 
which would establish an uniform lime throughout the Kingdom. 
He suggested that the Mayor should communicate with the Chief 
Magistrate of Devonporton the subject. 

Mr. W. F. Collier remarked it would be well if the Mayor could, 
at the same time, cause Railway trains to arrive at the hours at 
which they were due {hear, heur).* 7 

Legal time 

Although by 1855 98 per cent of the public clocks in Great 
Britain were set to GMT, there was still nothing on the 
statute book to define what was the time for legal purposes. 
Say someone died in Inverary at 11.50 p.m., just before 
midnight Sunday Greenwich time, but 10 minutes after 
midnight Monday Inverary time: did he die on Sunday or 
Monday? The answer to that question could have important 
legal consequences in insurance, inheritances, etc. The case 
of Curtis v. March, 25 November 1858, tested this matter: 

This was an action of ejectment which was entered for trial before 
Watson B | Baron of the Exchequer] at the last Dorchester Assizes. 
The time appointed for the sittingof the Court was 10 o'clock A.M., 
and the learned Judge took his seat on the bench punctually at 10 
by the clock in Court. The cause was then called on and the 
plaintiff's counsel commenced his address to the jury, but as the 
defendant was not present and no one appealed for him, the 
learned Judge directed a verdict for the plaintiff. The defendant's 
counsel then entered the Court and claimed to have the cause tried, 


ii 4 


on the ground that it had been disposed of before 100'clock. At that 
time it wanted one minute and a half to 10 by the town clock. The 
clock in Court was regulated by Greenwich time, which was some 
minutes before the time at Dorchester. . . . 5S 

On appeal, Chief Baron of the Exchequer Pollock, sitting 
with Barons Watson and Channell, reversed Watson's 
original decision: 

Ten o'clock is 10 o'clock according to the time of the place, and the 
town council cannot say that it is not, but that it is 10 o'clock by 
Greenwich time. Nor can the time be altered by a railway company 
whose railway passes through the place nor by any person who 
regulates the clock in the town-hall. 59 

The decision reached in 1858, specific to the time of sitting of 
courts, was held also to define legal time for other purposes 
in Great Britain until 1880: 'The time appointed for the 
sitting of a Court must be understood as the mean time at 
the place where the Court sits, and not Greenwich time, 
unless it be so expressed.' 6 " 

On 14 May 1880 The Times published a letter from 'Clerk 
to the Justices': 

During the recent elections many members of Parliament and the 
officials conducting elections must have been much troubled to 
decide what was the correct time to open and close the poll. 

Greenwich time is now kept almost throughout England, but it 
appears that Greenwich time is not legal time. For example, our 
polling booths were opened, say, at 8. 13 and closed at 4. 1 3 p.m. 

This point as to what is legal time often arises in our criminal 
courts, but has hitherto escaped a proper decision and discussion. 
Will not some new M.P. take up this point and endeavour to get an 
Act making Greenwich time legal time! 61 

On 1 June the same year the Statutes (Definition of Time) 
Bill was read for the first time in the House of Commons, 
sponsored by Dr Cameron, Mr David Jenkins, and Mr 
Ervington. It was referred to a committee whose report was 
laid before the House by Mr Playfair on 5 July. It passed 
through the remaining stages in both Houses almost with- 
out debate and received the Royal Assent on 2 August 
1880. 62 It stated firmly, and at last: 

Whenever any expression of time occurs in any Acts of Parliament, 
deed, or other legal instrument, the time referred shall, unless it is 


otherwise specifically stated, be held in the case of Great Britain to 
be Greenwich mean time, and in the case of Ireland, Dublin mean 

A story told about 1950 by Mr J. H. Garner, a super- 
intendent of the Central Telegraph Office, epitomizes the 
way in which the dissemination of a standard time system 
brought people together. After the Rugby time signal was 
instituted in 1927, the telegraph lines once used for time 
signals were 'recovered' so that they could be used for other 

Large numbers of lines were nominated as chronopher lines, with a 
number of rescues, and great care was always taken that all such 
nominated lines were tested and maintained so that they were all in 
good order for the time signal. One such reserve was the TS-BS 
London- Bristol - proved and checked every day - and regularly at 
9.0 a.m. the time signal was transmitted, as was thought, to a 
centre on the north coast of Somerset. . . . 

An enquiry was begun into whether the cost of maintenance of 
such lines was justified and in due course the Bristol line came 
under review, with the remarkable result that no trace of any circuit 
beyond Bristol could be found. Enquiries established that, years 
before, the line ran to a Customs and Excise and Coastguard centre 
on the Somerset coast, but changes had brought the centre into 
disuse and now only one coastguard remained. 

The engineers, being very thorough, traced and recovered what 
was left of the circuit and found its termination at the coastguard 
house where, boldly shown in the office window, was the well- 
known rod and ball with a notice that at 9.0 a.m. precisely the ball 
would fall to indicate Greenwich time. 

The coastguard admitted that he had not received a time signal 
for many years but the inhabitants expected a signal from him so he 
knocked the ball down with a stick each day. When asked how he 
got the time he said that by standing on a chair and looking along 
and across the road he could see by the 'Brown's' big clock when to 
knock the ball down. 

He was told to cease the practice and the engineers, being 
curious, went along to examine Brown's clock. They found it a 
well-built English clock, and only 15 seconds slow. Asked how he 
kept the clock at correct time, Mr. Brown said: 'Oh, that's easy. By 
standing on my stool and peering sideways along and across the 
road, I can just see in the coastguard's front window the brass ball 
fall on the rod.' 63 


O A prime meridian 1790-1884 

The longitude of observatories 

Before the era of railways and telegraphs, the only 
non-nautical requirement for Greenwich time to be known 
outside Britain was scientific - specifically geographic- in 
the precise measurement of the difference of longitude 
between observatories. Differenceof longitude can be deter- 
mined astronomically or geodesically (by trigonometrical- 
survey methods) or, as we have seen, by the transport of 
chronometers. One of the earliest examples of the use of this 
last method took place during the geodetic operation to 
connect Paris and Greenwich Observatories in 1784-8, 
instigated by Cassini de Thury and conducted on the 
English side of the Channel by Major-General William 
Roy, FRS. In September 1785 Maskelyne sent his assistant 
Joseph Lindley by post-chaise and cross-channel packet to 
Paris and back carrying eight of John Arnold's chron- 
ometers, yielding a difference of longitude of 9 minutes 19.8 
seconds, only about a second too small and agreeing well 
with the existing astronomical determinations and the geo- 
detic result. 1 In 1825 a series of rockets was used by John 
Herschel and Col. Sabine to connect Paris and Greenwich. 
The chronometer method continued to be used for longi- 
tude determination of observatories until the coming of the 
electric telegraph. For example, in 1843 more than sixty 
chronometers were sent sixteen times backwards and for- 
wards between Altona near Hamburg and Pulkowa near 
today's Leningrad, and the following year forty chron- 
ometers went the same number of times between Altona 
and Greenwich. Chronometers were sent across the 
Atlantic many times to determine the longitude difference 
between Harvard and Liverpool Observatories, from which 

A PRIME MERIDIAN 1790- 1884 

the difference of longitude between Harvard and Green- 
wich was accurately determined. In 1844 the longitude of 
Valentia Island and the west coast of Ireland was found in 
this way, under Airy's superintendence. 

The use of the electric telegraph for this purpose was first 
suggested by the American astronomer S. C. Walker and 
first used in the USA about 1849. As we have seen, the 
telegraphic connections between Greenwich and the 
Continent were suggested by Airy in 1851, connection with 
Brussels being established in 1853, and with Paris in 1854. 
The longitude of Valentia was redetermined by telegraph in 
1862. Telegraph signals do take a finite time to travel along 
the lines, depending upon the distance and the number and 
type of relays. Airy reported a time of passage of Via second 
from Greenwich to Paris in 1854 and '/ H second (nearly) to 
Valentia in 1862.- 


The Atlantic cable 

The first successful submarine cable was laid across the 
English Channel in 1851. Wales and Scotland were linked 
with Ireland in 1852, England with Belgium and Denmark in 
1853. By i860 London was connected with the Indian sub- 
continent, one of the longest submarine cables being be- 
tween Malta and Alexandria, 1,565 miles. But the really 
exciting prospect was a cable - perhaps more than one - 
between Europe and North America. Its main protagonist 
was the great American Cyrus W. Field (1819-92), whose 
untiring efforts provided the impetus throughout. 

The Atlantic Telegraph Company was formed in October 
1856 specifically to establish telegraphic communication be- 
tween Newfoundland and Ireland. In 1857 half the cable to 
be used was embarked in the US steam frigate Niagara at 
Birkenhead, the other half in HMS Agamemnon at the works 
of the Telegraph Construction and Maintenance Company 
in East Greenwich, within sight of the Royal Obser- 
vatory. After an unsuccessful attempt in 1857, a second was 
made in the spring of 1858. On 5 August the first telegraph 
message was sent across the Atlantic, causing great jubila- 
tion on both sides of the ocean. However, by 3 September, 
less than a month later, communication failed and could not 
be re-established. 


The American Civil War between 1861 and 1865 brought 
only a temporary halt in Field's efforts to achieve a trans- 
atlantic cable. More capital was raised and heavier cable 
manufactured at Greenwich. For two ships to lay the cable 
had not proved entirely satisfactory, so it was decided to 
use the only ship then capable of embarking all the cable 
needed - the Great Eastern, 22,500 tons, the conception of 
Isambard Kingdom Brunei, built at Millwall on the Thames 
and the world's largest ship. Because of her draught (nearly 
35 feet when loaded) the Great Eastern had to lie at 
Sheerness to embark the cable, which had been sent down- 
river by lighter from Greenwich. 

37 On 23 July !865 she left Valentia, Ireland, paying out her 
cable, escorted by 1 IM Ships 'terrible and Sphinx and having 
on board Professor William Thomson, later Lord Kelvin 
(1824-1907), one of the greatest names in Victorian science. 
Alas, on 2 August, after 1,025 nautical miles of cable had 
been paid out, it parted and could not be recovered. The 
next year, however, a new cable was successfully laid by the 
Great Eastern, taking fourteen days from Valentia to Heart's 
Content in Newfoundland. To complete the triumph, the 

38 Great Eastern successfully grappled the 1865 cable, spliced it 
on to cable remaining on board, and thus provided a second 
cable link across the Atlantic. 

One of the factors leading to this success was that during 
the 1866 lay, at the suggestion of Captain Anderson, the 
Greenwich time signal was received by the Great Eastern 
twice daily by telegraph via London, Holyhead, Dublin, 
Valentia, and the cable she was laying, thus enabling her to 
find her longitude exactly.- 1 This seems to be the earliest 
example of a ship at sea receiving a time signal by other than 
visual means. One of the earliest uses of the new cable was 
to redetermine the longitude difference between the 
observatories of Greenwich and Harvard University at 
Cambridge, Mass. This was conducted in October 1866 by 
Dr B. A. Gould of the US Coast Survey, in co-operation 
with Airy. 4 

European railways 

On the continent of Europe, the railways brought the same 
problems of timekeeping as they had in Britain. Generally 


speaking, though local mean time was kept by the passen- 
gers, the trains in each country were run according to some 
central time. In France, for instance, clocks inside railway 
stations were kept to I'lieure de la gare which was 5 minutes 
slow on Paris time, while clocks outside the station were 
kept to local time, I'lieure de la ville. Belgian trains ran 
to Brussels time, Dutch trains to Amsterdam time. In 
Germany railway officials kept any one of five times - those 
of Berlin, Munich, Stuttgart, Karlsruhe, or Ludwigshafen. 
Passengers, however, kept strictly to local time and then' 
were posts set alongside the rails marking each minute's 
change of time. It was apparently customary for watches to 
be altered in ten-minute steps during the journey. This 
fixation on local time, encouraged by the German astron- 
omers, was only abandoned, and Berlin time adopted 
throughout, when Count von Moltke pointed out the 
military consequences of this lack of standardization. 5 AH of 
which must have been very confusing to long-distance 

US railroads 

If the need for co-ordination of time-keeping was evident in 
Great Britain with a maximum longitude difference equiva- 
lent to 30 minutes of time, how much greater were the 
potential difficulties in the United States where the differ- 
ence between east and west amounted to more than 3V2 
hours? However, it was not this longitude difference that 
drew attention to the problem, but the fact that the many 
railway companies which sprang up after the end of the 
Civil War each kept its own time, as did each town and city 
on the way, to the great inconvenience of the travelling 
public. For instance, a traveller from Portland, Maine, on 
reaching Buffalo, NY, would find four different kinds of 
'time': the New York Central railroad clock might indicate 
12.00 (New York time), the Lake Shore and Michigan 
Southern clocks in the same room 11.25 (Columbus time), 
the Buffalo city clocks 11.40, and his own watch 12.15 
(Portland time). At Pittsburgh, Penn., there were six differ- 
ent time standards for the arrival and departure of trains. A 
traveller from Eastport, Maine, going to San Francisco, was 
obliged, if anxious to have correct railroad time, to change 

A PRIME MERIDIAN 179O- 1884 121 

his watch some twenty times during the journey. 6 

The first observatory in the USA to distribute time seems 
to have been the Naval Observatory in Washington, from 
August 1865. 7 Then in 1869 Professor S. P. Langley of the 
Allegheny Observatory near Pittsburgh instituted a distri- 
bution system which covered a very considerable area, from 
Philadelphia and New York in the east, Lake Erie in the 
north, and Chicago in the west. But three separate time 
signals were necessary - at Pittsburgh time for the local 
watchmakers and jewellers; at Altoona time (10 minutes 
fast on Pittsburgh) for the Pennsylvania Central Railroad to 
Philadelphia and eastwards; and at Columbus time (13 
minutes slow on Pittsburgh) for the Pittsburgh, Fort 
Wayne, and Chicago railroad. 8 In his description of the 
system, Langley put in the strongest plea for a uniform 
standard of time, noting that the managers of the railroads 
connecting New York, Philadelphia, Pittsburgh, and 
Chicago were then (1872) contemplating the use of one 
standard time, that of the meridian of Pittsburgh, for all 
trains between the cities concerned. 

Meanwhile, there were other thoughts about the matter 
of standardizing time for railroad and other purposes in the 
United States. In 1870 a 107-page pamphlet was published 
entitled A System of National Time for Railroad*, written 
by Professor Charles Ferdinand Dowd (1825-1904), 39 

Principal of Temple Grove Ladies' Seminar)' in Saratoga 
Springs, NY. Dowd's pamphlet was a result of discussions 
at the Convention of Railroad Trunk Lines in New York City 
in October 1869. To obviate the inconvenience of having 
some eighty different time standards on the various US 
railroads - not to mention the fact that each station had to 
maintain its own local time as well - Dowd proposed a 
scheme identical in principle with the standard time system 
used all over the world today: that, for Railroad Time pur- 
poses in the USA, there should be four standard meridians 
15 (= 1 hour) apart, the eastern one being the Washington 
meridian. These meridians would be the centres of four 
time zones; in each zone the time adopted would be 
uniform, and it would change by one hour when passing 
from one zone to the next. The boundaries of each zone, 
though approximating to the appropriate meridians, were 

39- Charles P. Dowd, of Saratoga Springs, NY. who first devised the /one-time system used all 
over the world today. From Harper's Weekly, 29 Dec. 1883. 

A PRIME MERIDIAN 1790 - 1884 

to be adjusted to take into account local state or county 
boundaries and the areas served by individual railroads. 
This meant that, in theory, the minute and second hands of 
all clocks would show the same time, only the hour hands 
being different. But Dowd's first scheme differed from the 
zone-time system of today in one particular - it was based 
on the meridian of Washington, specifically that passing 
through the transit instrument at the US Naval Observa- 
tory. (It is of interest that, in the Act of 28 September 1850 
making the appropriation for the US Nautical Almanac, 
Congress made the proviso 'that hereafter the meridian of 
the observatory at Washington shall be adopted and used as 
the American meridian for all astronomical purposes, and 
that of the meridian of Greenwich shall be adopted for all 
nautical purposes'.) 

The convention agreed wholeheartedly in principle with 
Dowd's proposals and he was asked to go and work out the 
scheme in detail. For instance, it was obviously impracti- 
cable for the limits of each zone to conform exactly to the 
appropriate meridian or one could have different times 
being kept in different rooms of the same house. To con- 
tinue the story in Dowd's own words: 

I at first took the national meridian of Washington, and having 
divided the country into three 15-degree belts, patiently marked 
out the longitude of some 8,000 stations along some 500 railroad 
lines, and had a map engraved showing the hour sections and the 
proposed Standard versus the actual time at each station. With this 
map was incorporated and published a pamphlet of 100 octavo 
pages, which I sent to all railroad men and others in this country 
who would likely feel an interest in the work. 

In 1871 I presented the subject to the North East Railroad Asso- 
ciation in Boston, and to others elsewhere. Then came suggestions 
that brought to my mind that it would be better to adopt the 
nautical meridian of Greenwich, and by using its fifth hour or the 
75th meridian west longitude as the Prime American standard, I 
laid out the hour sections upon that basis. In the spring of 1872 I 
went to St. Louis and attended a meeting of the Western Railway 
Association, and to Atlanta, Georgia, to confer with super- 
intendents of Southern railways.* 

Apparently Dowd learned from these meetings that using 
the Washington meridian projected the other meridians too 
far west to suit the needs of the Eastern and Central 



sections, which felt their convenience and that of 'the 
borderline cities' should be considered. So, in May 1872, 
Dovvd changed the system he was promoting and moved all 
the hour sections approximately two degrees east so that 
they now conformed with integral numbers of hours west of 
the Greenwich meridian. He summarized his proposals as 

Explanation of standards 

The time of the 75th Meridian [west of Greenwich] is adopted as 
the standard time for all roads east of Ohio and the Allegheny 
Mountains; and the time of the 90th Meridian for western roads 
situated anywhere in the Mississippi valley. These times may be 
designated Eastern and Western times, their difference being just 
one hour. Following westward still, the next hour standard falls in 
the Rocky Mountain District, and hence is of little avail. But the 
third hour standard, or the time of the 120th Meridian, is very 
central and convenient for roads on the Pacific coast. Again the 
fifth hour eastward is adopted as the standard time of England, and 
is the basis of longitude on all marine charts. 

Saratoga Springs, N.Y.. C. F. Dowd. 1 " 

May 15, 1872. 

As might be expected, Dowd's proposals gave rise to 
much debate, and many alternative suggestions were put 
forward, in particular those of 1879-80 where the use of a 
single standard time, that of the 90th meridian west of 
Greenwich (which almost passes through St. Louis and 
New Orleans), was seriously considered. The full story is 
too long to give here and has in any case been lucidly 
summarized elsewhere." Eleven years and many railroad 
conventions later, Dowd's 1872 plan was adopted virtu- 
ally unchanged by the railways of the USA and Canada. At 
noon or before on Sunday 18 November 1883 public clocks 
all over North America were altered to the 'new standard of 
time agreed upon, first by the railroads, for the sake of the 
uniformity of their schedules, but since generally adopted 
by the community through the action of various official and 
corporate bodies as an obvious convenience in all social and 
business matters'. 12 

Newspapers all over the country reported reactions. The 
Neio York Herald, for example, writing on change-over day, 
pointed out that someone going to church in New York that 


day would discover that the noon service had been curtailed 
by almost four minutes, while every old maid on Beacon 
Hill in Boston would rejoice that night to discover that she 
was younger by almost sixteen minutes. On the other hand, 40 



. Standard Time zone boundaries 
adopted by railroads Nov 18. 1883 

■ Present Standard Time zone boundaries 

40. US time zones, in i88vnd today. After C. 1 Corliss, TheDu/ofltonNoons (Washington, 


everyone in Washington would be eight minutes older. In 
the central zone, the new time in Chicago was 9'^ minutes 
slower than the old; in Cleveland, Ohio, not far from the 
border of the zone, 32 minutes slower. The next day, in an 
article headlined 'Horometric Harmony', the Herald re- 
ported that great crowds had watched the dropping of the 
time-ball on the top of the Western Union building. At the 
Church of Our Father in Brooklyn a sermon was preached - 
duly reported in about six column-inches - on 'Changing 
and Keeping Time', with the text from Joshua 10: 13: 'And 
the sun stood still, and the moon stayed, until the people 
had avenged themselves upon their enemies. Is not this 
written in the book of Jashar? So the sun stood still in the 
midst of heaven, and hasted not to go down about a whole 
day-' 13 



Harper's Weekly, besides reproducing a portrait of Dowd 
and a map illustrating the new time zones, outlined the 
advantages in the flowery prose of the 1880s: 

On Saturday, the 17th of November, when the sun reached the 
meridian of the eastern border of Maine, clocks began their jangle 
for the hour of twelve, and this was kept up in a drift across the 
continent for four hours, like incoherent cowbells in a wild wood. 

But on Monday, the 19th - supposing all to have changed to the 
new system on the 18th - no clock struck for this hour till the sun 
reached the seventy-fifth meridian. Then all the clocks tin the 
continent struck together, those in the Eastern Section striking 
twelve, those in the Central striking eleven, those in the Mountains 
striking ten, and those in the Pacific striking nine. 

The minute-hands of all were in harmony with each other, and 
with those of all travellers' watches. Time-balls everywhere be- 
came perfectly intelligible, and the bliss of ignorance was no longer 
at a premium. IJ 

So, in 1883, the United States and Canada adopted a time 
standard based upon the Greenwich meridian. Although it 
was not until 1918 that an Act of Congress legalized 
standard time all over the United States, the civil population 
nevertheless adopted 'Railroad Time' almost spon- 
taneously, as had happened in Britain thirty years before: 85 
per cent of US towns of over ten thousand inhabitants had 
done so by October 1884. There were exceptions. Detroit, 
Michigan, for example, on the borderline between the 
Eastern and Central zones, continued to keep local time 
until 1900 when the City Council decreed that clocks should 
be put back twenty-eight minutes to Central Standard 
Time. Half the city obeyed, half refused. After considerable 
debate, the decision was rescinded and the city reverted to 
Sun time. A derisive offer to erect a sundial in front of the 
city hall was referred to the Committee on Sewers. IS Then, 
in 1905, Central time was adopted by city vote. In May 1915 
this was changed by ordinance to Eastern Standard, a 
decision upheld by popular vote in August 1916. '• 

Though, as we shall see, many others - particularly the 
Canadian, Sandford Fleming - should be given the credit 
for encouraging the adoption of Standard Time on a world- 
wide basis, it is to Charles F. Dowd that we are indebted for 
the principle. Ironically, he met his death beneath the 

A PRIME MERIDIAN 1790- 1884 

wheels of a railroad locomotive in Saratoga, New York, in 

The prime meridian 

Meanwhile, particularly since 1870, geographers and scien- 
tists of allied disciplines of all nations had been giving their 
attention to the possibility of fixing a common zero for 
longitude and time-reckoning throughout the globe. And it 
was the former- the prime meridian - which was the first to 
be discussed. As we have seen, the first astronomer to 
determine differences of longitude seems to have been 
Hipparchos. For his first meridian - or prime meridian, or 
miridwn-origine, or meridien initial, or meridiaiio initiate, or 
Nulhneridien - he used Rhodes, where he was observing. 
But Ptolemy, following Marinus of Tyre, adopted a mer- 
idian through the Insulae Fortunatae, the Canary Islands, 
which seemed to mark the western boundary of the world, 
whereas, to the east, there seemed to be no such boundary. 

With the voyages of discovery in the fifteenth and six- 
teenth centuries, a new interest arose. In 1493 Pope 
Alexander VI laid down a line of demarcation between the 
spheres of influence of Portugal and Spain 100 leagues west 
of the Azores and the Cape Verde Islands; after protests by 
Portugal, this was shifted, in the Treaty of Tordesillas 
(1494), to a meridian 370 leagues west of the Cape Verde 
Islands which was itself used on some maps and charts as a 
prime meridian. On 13 July 1573 Philip II of Spain issued 
two Ordinances concerning the measurement of longitude 
throughout the Spanish empire. The first of these (N0.62) 
stipulated that all longitudes should be computed from the 
meridian of the city of Toledo and, contrary to the practice of 
the ancients who measured eastwards (from the Canaries), 
Spanish longitudes should be measured westwards 'be- 
cause to proceed in this manner is more natural and con- 
forms to the discovery of the Indies which God was pleased 
to grant us'. 17 Ordinance No. 67 ordered all colonial 
governors to take every opportunity to have eclipses and 
other phenomena capable of yielding difference of longi- 
tude observed whenever possible, the results to be reported 
to the Chief Cosmographer of the Casa de Contratacion 
(House of Trade) at Seville, so that the longitudes of places 



could be computed to complete the Padtvn Real, the com- 
plete inventory of facts and data about the New World 
which made possible the plotting of the official Spanish 
nautical charts. 

Later in the sixteenth century cartographers such as 
Mercator and Ortelius began to choose various islands in 
the western ocean - in the Canaries, Madeira, the Cape 
Verdes, and even the Azores - for reasons described 
vividly by William Blaeu in the Latin cartouche of his globe 
of 1622: 

... But in our days a good many think this starting point ought to 
be based on nature itself, and have taken the direction of the 
magnetic needle as their guide and placed the prime meridian 
where that points due north. But that these are under a delusion is 
proved by that additional property of the magnetic needle through 
which it is no standard tor the meridian, for itself it varies along the 
same meridianaccordingasitisnearone land mass or another. . . . 

Blaeu goes on to explain that, for his own globe, 

. . . following in the steps of Ptolemy, |we| have chosen the same 
islands and in them Juno, commonly called Tenerife, whose lofty 
and steep summit covered with perpetual cloud, called by the 
natives £.7 Pico, shall mark the prime meridian. In that way we have 
differed barely a quarter of a degree from the longitude of the Arabs 
who chose the extreme western shore of Africa |C. Verde), and 1 
thought it well to point this out. Is 

In April 1634 Cardinal Richelieu called a conference of 
eminent European mathematicians and astronomers in 
Paris to consider the question of a prime meridian to be 
recognized by all nations. The choice fell upon Ptolemy's 
Fortunate Isles, more closely defined as the west coast of the 
island of Ferro (or He de Fer, or Hierro), the westernmost of 
the Canary Isles. However, the Thirty Years War was in 
progress and, judging by the terms of Louis XIII's decree of 
i July 1634, the motive behind the calling of the conference 
seems to have been at least as much political as scientific: 

French ships are not to attack Spanish or Portuguese ships in 
waters lying east of the First Meridian and north of the Tropic of 
Cancer. In order that this first meridian may be more clearly known 
than it has for some time been, the Admiral of France has consulted 
persons of knowledge and experience in navigation. The King in 


consequence forbids all pilots, hydrographers, designers or en- 
gravers of maps or te r re s t ri al globes to innovate or vary from the 
ancient meridian passing through the most westerly of the Canary 
Islands, without regard to the novel ideas of those who recently 
fixed it in the Azores on the supposition thai there the compass 
does not vary, for it is certain that this happens also in other places 
that have never been taken lor the meridian. ''' 

In 1724 Louis Feuillee was sent by the Paris Academy to 
make actual measurements to determine the longitude of 
Paris based on Richelieu's prime meridian. Depending 
upon Feuillee's astronomical observations taken on the 
neighbouring island of Tenerife, the figures eventually 
published in 1742 were: Paris (Notre Dame) 20° 02'. 5 east of 
the western point of Ferro; London (St. Paul's) \f 37'. 5 E. 2 " 
1 lowever, there was still no general agreement on a prime 
meridian and each nation tended to please itself, generally 
using its capital city or principal observatory. Navigators in 
their logs generally used the point of departure on any 
specific leg of the voyage, 15 27' west of the Lizard, or 
26 32' east of the Cape of Good I lope, for example. Most 
sea charts in the eighteenth century had only a single longi- 
tude scale (if any at all) whose origin depended upon the 
national origin of the chart. The French, however, logical as 
always, put multiple longitude scales on many of the charts 
in the great series of official sea-atlases, Neptune Francois, so 
that the navigator could at will plot his position relative to 
Tenerife, Ferro, the Lizard, and Paris. Incidentally, the 
same charts also had multiple scales of distance - of Breton 
leagues, French leagues, English leagues, as well as nautical 
miles - symptomatic of the confusion that prevailed before 
the days of international standardization of weights and 

We have seen that the publication of the British Nautical 
Almanac in 1767 meant that Greenwich began to be used as 
the prime meridian on maps and charts of many nations 
from the late 1700s. The first series of charts to use 
the Greenwich meridian systematically was J. F. W. 
DesBarres's Atlantic Neptune covering the east coast of 
North America from Labrador to the Gulf of Mexico, first 
published in 1784. DesBarres's charts continued to be used 
as the primary source for most American charts for the next 



■ ■ /" — '■■' /' — i — ia 
gipl9<gpt(i>ln>l*lr ilu vf,..l', r ., J. | 

41 . Multiple longitude scales on the charts in the official French atlases, Lt Neptune (VflMfliis. Det.iil 
from 'Carte p.irticulieredes Cosies de Bretagne', c, 1771 from an original of 1693. 


fifty years and, with the British Nautical Almanac itself, were 
probably the primary reason for the decision by the US 
Government to retain the Greenwich meridian for nautical 
purposes in 1850. In 1853 the High Admiral of the Russian 
Fleet cancelled the use of the nautical almanac specially 
prepared for Russia and, in its place, introduced to the 
Russian Navy the British Nautical Almanac, based on the 
Greenwich meridian, from which the Morski Miesiatseslob 
(Naval Almanac) was produced. 21 

International discussions 

The first International Geographical Congress (1GC) took 
place at Antwerp in August 1871. One of the resolutions 
passed expressed the view that, for passage charts of all 
nations (but not necessarily coastal or harbour charts), the 
Greenwich meridian should be adopted as the common 
zero for longitude, and that this should become obligatory 
within fifteen years. It was also recommended that, when- 
ever ships exchanged longitudes at sea, they should be 
based on Greenwich. For land maps and coastal charts, 
however, each state should keep its own prime meridian. 22 
In the discussion, M. Levasseur, one of the French repre- 
sentatives, generously said that, had it been the seven- 
teenth or eighteenth century, the choice must have fallen on 
Paris, but that circumstances had changed: the majority of 
charts used at sea were now of British provenance and, 
furthermore, le livre habitue! du matin was the British Nautical 
Almanac. (A few years later, it was said that the annual sales 
of the Nautical Almanac were 20,000 copies; of the 
Connaissance des Temp*, 3,000 copies.) 23 M. Levasseur there- 
fore supported the resolution as far as sea charts were 

The 2nd IGC in Rome in 1875 discussed the whole matter 
again without coming to any further conclusions. France, 
however, expressed a sentiment which was to be renewed 
time and again in subsequent discussions - that if England 
were to accept the metric system, then it would perhaps be 
courteous of France to accept the Greenwich meridian. 24 
Table I shows that, only a dozen years later, the 1871 resolu- 
tion had begun to take effect: twelve nations were counting 
their longitudes from Greenwich on newly published sea 


1 3 2 



charts. On land maps, however, each nation continued to 
go its own way. 

In 1876 there came a development from across the 
Atlantic, when a memoir called Terrestrial Time was pub- 
lished in Canada by Sandford Fleming (1827-1915), 
engineer-in-chief of the Canadian Pacific Railway. Born in 
Scotland, where he studied engineering and surveying, 
Fleming went to Canada in 1845. He started his career as a 
railway builder, becoming chief engineer of the Ontario, 
Simcoeand Huron (later Northern) Railway in 1852. In 1862 
he took a similar post with the Intercolonial Railway and 
was appointed engineer-in-chief of the Canadian Pacific 
Railway, where he surveyed the Yellowhead Pass route 
(now followed by the Canadian National Railway) and was 
the first to demonstrate the practicability of the route 
through the Kicking Horse, Eagle, and Rogers Passes. 
Originally Fleming was interested only in the use of a 24- 
hour clock system, but then became fired with enthusiasm 
for the idea of a uniform time for the whole world - 
Terrestrial Time, or, as he later suggested, Cosmopolitan 
(and later still Cosmic) Time. He suggested an hour-zone 
system similar to Dowd's for use as local time for domestic 
purposes (he did not acknowledge Dowd's idea, though, 
being in the railroad business, he must have known about 
it), while his Terrestrial Time would be used by railroads, 
telegraphs, the sciences, etc. 2s Later, Fleming was less 
enthusiastic about Dowd's standard-time system. 26 

In 1878-9 Fleming read two papers before the Canadian 
Institute, Toronto, entitled respectively 'Time-reckoning' 
and 'Longitude and time-reckoning'. The first was a re- 
written version of his 1876 paper; the second, subtitled 'A 
few words on the selection of a prime meridian to be 
common to all nations, in connection with time-reckoning', 
put a strong case for the prime meridian being 180° from 
Greenwich, coinciding with the basic meridian used for 
today's International Date Line. 27 Fleming's two papers 
were considered so important that in June 1879 the British 
Government forwarded copies to eighteen foreign coun- 
tries and to various scientific bodies in England. 



42. Sir Sandford Fleming. 



Table I Prime meridians in use in the early 1880s on newly 
published maps and charts 

Prime meridian 



Sen charts 

/ and maps 




Rio de Janeiro 

Greenwich and 

Rio de Janeiro 

Denmark Greenwich, Copcn- Copenhagen 

hagen, and Paris 

France and Algeria Paris Paris 

Germany Greenwich and I erro Ferro 

I lolland Greenwich Amsterdam 

India Greenwich 

Italy Greenwich Rome 

lapan Greenwich Greenwich 

Norway Greenwich and Ferro and Christiania 


Portugal Lisbon Lisbon 

Russia Greenwich, Pulkowa, Ferro, Pulkowa, 

and Ferro Warsaw, and Paris 

Spain Cadi/ (S. Fernando) Madrid 

Sweden Greenwich, Stock- Ferro and Stockholm 
holm, and Paris 

Switzerland Paris 

UK and colonies Greenwich Greenwich 

USA Greenwich Greenwich and 


Sources See efforts: BORSARI, F., // meridkmo inhkde t I'om unmersele 
(Napc4i, 1883), 60. Land maps: WHEELER, CM.. Report on Hie Third Into 
national Geograplikal Congress. . . Venice. . . 18S1 (Washington, 1885). 30. 

On the whole, reaction was favourable. However, there 
were some notable exceptions, particularly among astron- 
omers. Sir George Airy, writing in 1879, two years before 
his retirement, said that, first, he 'set not the slightest value 
on the remarks extending through the early parts of Mr. 
Fleming's paper' (about the adoption of hour zones for local 
time and the establishment of 'Terrestrial Time']; and 
secondly, 'As to the need of a Prime Meridian, no practical 
man ever wants such a thing. If a Prime Meridian were to be 
adopted, it must be that of Greenwich, for the navigation of 


almost the whole world depends on calculations founded 
on that of Greenwich. ... But I as Superintendent of the 
Greenwich Observatory, entirely repudiate the idea of 
founding any claim on this: Let Greenwich do her best to 
maintain her high position in administering to the longitude 
of the world, and Nautical Almanacs do their best, and we 
will unite our efforts without special claim to the fictitious 
honour of a Prime Meridian.' 28 At Airy's suggestion, the 
Governor-General was told on 15 October 1879 that the 
British Government would not interfere in a matter which 
concerned social usages. 29 It seems strange that a man with 
such vision - to whom, more than anyone else, the 
country was indebted for the dissemination of Greenwich 
time - should hold such views at that date: only a year later, 
GMT was to become legal time in Great Britain; only three 
years later, a time-zone system was to be introduced 
throughout Canada and the USA; only five years later, it 
was agreed internationally that a Prime Meridian ieas 
needed, and that it should be Greenwich. 

But among astronomers Airy was not alone. Professor 
Pia/./.i Smyth, Astronomer Royal lor Scotland, while prais- 
ing Fleming for his good intentions, condemned him for 
want of practicality. If there had to be a common prime 
meridian, why not the Great Pyramid in Egypt? 30 Professor 
Simon Newcomb, Superintendent of the American nautical 
almanac, was even more scathing. Asked in 1882 whether it 
was advisable for the USA to adopt a time system which 
would commend itself to other nations and be adopted by 
them ultimately, he answered, 'No! We don't care for other 
nations; we can't help them, and they can't help us.' To a 
second question as to whether the scheme for regulating 
time seemed to possess any features to commend itself to 
him, he said: 'A capital plan for use during the millennium. 
Too perfect for the present state of humanity. See no more 
reason for considering Europe in the matter than for con- 
sidering the inhabitants of the planet Mars.' 31 

After which, it is almost a relief to note the reactions of the 
astronomers of Bologna who suggested Jerusalem as the 
prime meridian; and of Professor de Beaumont, a geogra- 
pher from Geneva, who suggested a meridian straight 
through the Bering Straits, chosen so as to be an even 



number of degrees from both Paris and Ferro, with an 
anti-meridian (i8o°away) which would pass close to Rome, 
Venice, and Copenhagen." About this time also, the de- 
sirability of decimalizing the right angle and the hour (as 
had been originally conceived in the metric system) began 
to be discussed once again. All these views were in front of 
the delegates to the Third International Geographical Con- 
gress, which met in Venice in September 1881 and for which 
the establishment of a universal prime meridian and a uni- 
form standard of time was high on the agenda. There was 
considerable discussion on all these matters but the impor- 
tance of this congress in the present context lay mainly in 
what stemmed from it - the two special conferences about 
to be described." 

The first of these was the Seventh International Geodesic 
Conference which assembled in Rome in October i883-This 
conference of men of science is particularly important in 
that its conclusions formed the basis of the subsequent 
Washington conference which was primarily diplomatic in 
character. The Rome conference was attended by astron- 
omers, geodesists, and mathematicians. Great Britain and 
the USA were specially invited to send representatives in 
view of the main subject matter - longitude and time. 
Britain was represented by W. H. M. Christie (1845-1922) 
who had succeeded Airy as Astronomer Royal two years 

They began their discussions with admirable dispatch, 
the scientific objectivity, practical approach, and lack of 
national prejudice being particularly noticeable. 'The time 
has passed when pure science thought it beneath its dignity 
to concern itself when necessary with matters of general 
practical usefulness, and when governments and admini- 
strations thought they could manage without consulting 
men of learning on questions relating to science.' So ran the 
official report, which continued by noting that, though the 
conference was not empowered to make final decisions or 
definitive international agreements, nevertheless by its 
constitution it was likely to have an important influence on 
future decisions. And so it proved. 

Resolution I epitomized the objective attitude of the dele- 
gates, stating that the unification of longitude and time was 


desirable as much in the interest of science as of navigation, 
commerce, and international communications, and sug- 
gesting that the importance of such a measure far out- 
weighed any sacrifices which might have to be made. 
Resolution II recommended the extension of the decimal 
division of the right angle to geodesic calculations, while 
retaining the sexagesimal system (the traditional degrees, 
minutes, and seconds) for astronomy, navigation, maps 
and so on. 

Resolution III was the fundamental one, concerning the 
actual selection of the prime meridian. In the absence of a 
natural zero for longitude (as the equator is the natural zero 
for latitude), the need for a prime meridian to be chosen 
arbitrarily was stressed and the scientific conditions to be 
fulfilled by such a meridian were discussed at length. Ideas 
of basing the new prime meridian on Ferro, on 20 west of 
Paris (which is nearly Ferro), on the Bering Straits, or of 
finding a 'non-natural' (or neutral) meridian - all these were 
dismissed. The initial meridian must be defined by an ob- 
servatory of the first order and, after noting that 90 per cent 
of navigators engaged in foreign trade already calculated 
their longitudes from Greenwich, the Conference proposed 
that Governments should adopt Greenwich as the initial 
meridian. Resolution IV, saying that longitudes should be 
counted in a single direction (fromo° to 360 ), seems to have 
slipped through with almost no debate and was to be 
negatived the following year in Washington. 

Having disposed of longitude, the conference then went 
on to discuss the unification of time, recognizing in Resolu- 
tion V the usefulness of adopting a universal time for certain 
scientific needs and for internal use in railways, shipping 
lines, telegraphs and posts, to be used alongside local or 
national time which must necessarily continue to be used in 
civil life. (The USA and Canada adopted Standard Time 
only a month later.) In Resolution VI the conference 
recommended that the Universal Day should start at 
Greenwich noon, so that the new Universal Day and the 
Astronomical Day could be reconciled. This recommenda- 
tion was also reversed at Washington. 

In winding up, the conference hoped that, if the entire 
world was prepared to accept Greenwich as the prime 



meridian, Greal Britain on her part might be prepared to 
conform to the metric system. It recommended a very early 
international convention devoted to the unification of longi- 
tude and time 'such as the United States government has 
proposed'. 34 

The International Meridian Conference, Washington, 
October 1884 

As a result of the Venice geographical conference 
in September 1881, the United States passed an Act of 
Congress on 3 August 1882 authorizing the President to 
call an international conference to fix on a common prime 
meridian for time and longitude throughout the world. 35 
On 23 October 1882 the State Department sent a circular 
letter to their representatives abroad asking whether such a 
conference would be welcome. The circular explained that, 
as the USA possessed 'the greatest longitudinal extension 
of any country traversed by railway and telegraph lines', it 
was very appropriate that she should call the conference. 
The response to theUnited States proposals for a conference 
in Washington was very favourable and the need for it had 
been underlined by the Rome conference of October 1883. 
On 1 December, therefore, invitations were sent to all 
nations in diplomatic relations with the US to send dele- 
gates - not exceeding three - to a conference to assemble in 
Washington on 1 October 1884. In view of future decisions, 
it is worth remarking that the North American railroads had 
adopted a standard time system based on the Greenwich 
meridian only eighteen days before the invitations were 
sent out. 

On 1 October 1884 forty-one delegates from twenty-five 
countries assembled in Washington for the International 
Meridian Conference. (These figures do not include the 
delegate from Denmark who never arrived.) The full list of 
delegates is given in Appendix VI from which it will be seen 
that the majority were professional diplomats, though some 
countries sent scientific and technical representatives also. 
The conference was opened by the Secretary of State, the 
Hon. F. T. Frelinghuysen, in the name of President Arthur. 
The first business was to elect officers, Admiral Rodgers, 
USN, being elected President of the Conference, General 


Strachey of Great Britain, Prof. Janssen of France, and Dr 
Cruls of Brazil being elected as secretaries. 

The discussion on the main points then began, the aims of 
the conference having been defined in the Act of Congress: 
'. . . for the purpose of discussing, and, if possible, fixing 
upon a meridian proper to be employed as a common zero 
of longitude and standard of time-reckoning throughout 
the whole world . . .'. Over a period of a month the confer- 
ence met on eight occasions and the report of their proceed- 
ings occupies over two hundred pages. In the following 
summary the sub-titles quoted give the wording of the 
resolutions OS finally agreed, amendments and resolutions 
considered but not adopted being mentioned in the text if 
their importance warrants it. The voting country by country 
is analysed in Table II (pp. 146-7). 

"That it is tlie opinion of this Congress it is 
desirable to adopt a single prime meridian for all 
nations, in place of the multiplicity of initial 
meridians which nmv exist.' 

As a result of a suggestion from the Spanish delegate that 
discussion should be based upon the resolutions of the 
Rome conference, the conference proper opened with a 
resolution similar to Resolution II below - that Greenwich 
should mark the prime meridian - proposed by a US dele- 
gate. However, M. Lefaivre, a French diplomat, and his 
astronomical colleague, Prof. Janssen, suggested that this 
was prejudging the issue. As a result Cdr. Sampson of the 
USA proposed Resolution I which, after very little dis- 
cussion, was adopted unanimously. 


'That the Conference proposes to the Govern- 
ments here represented the adoption of the mer- 
idian passing through the centre of the transit 
instrument at the Observatory of Greenwich as 
the initial meridian for longitude.' 

The USA then once again proposed Resolution II. Prof. 
Janssen said that France was of the opinion that the purpose 
of the conference was to examine the principles upon which 



a prime meridian should be chosen, leaving the actual 
choice to a more technical conference. This provoked 
considerable discussion, the general view being that the 
whole purpose of the conference was not just to establish 
principles but actually to fix a prime meridian - as stated 
in the Act of Congress which initiated the conference. 
M. Lefaivre thereupon proposed another resolution - that 
the initial meridian should have 'a character of absolute 
neutrality. It should be chosen exclusively so as to secure to 
science and to international commerce all possible advan- 
tages and in particular especially should cut no great 
continent - neither Europe nor America.' 

Discussion on this last resolution was long and went on 
for the whole of the session and well into the next. How 
could a meridian be absolutely neutral, argued Great Britain 
and the United States? It was very important that the prime 
meridian should pass through an astronomical observatory 
of the first order. Modern science, said Captain Evans 
(Great Britain), demanded such precision that one must 
abandon all ideas of establishing the meridian on an island 
(the Azores and Ferro had been suggested), on the summit 
of a mountain (say Tenerife), in a strait (say the Bering 
Strait), or as indicated by a monumental building (the Great 
Pyramid and the Temple of Jerusalem had both been put 
forward previously). This really left only the observatories 
44 of Paris, Berlin, Greenwich, and Washington as satisfying 

scientific needs. Sampson (USA) added that it was impor- 
tant also 'to so fix and define it that natural changes of time 
may not render it in the least degree uncertain'. To define it 
as a certain number of degrees east or west of an established 
observatory did not thereby make it a neutral meridian, but 
merely disguised it. Surely one must also consider the prac- 
tical point of view: with a new neutral meridian, everyone 
would have to change. 

There was much discussion also on the metric system 
which the French suggested was a truly neutral system, 
'. . . and we are still awaiting the honour of seeing the 
adoption of the metrical system for common use in 
England'. No, said Professor Abbe (USA), though the USA 
and England used the metric system as the standard in all 
important scientific work, it was nevertheless not entirely 

A PRIME MERIDIAN 1790- 1884 

neutral, having been fixed by French measurement. 'Had the 
English, or the Germans, or the Americans taken the ten- 
millionth part of the quadrant of the meridian, they would 
have arrived at a slightly different measure. ... It was 
intended to be a neutral system, but it is a French system.' 

And so it went on, the same arguments being put forward 
again and again, the earlier discussions being exclusively 
conducted bv France on the one hand and the USA and 
Great Britain on the other - the one saying that the prime 
meridian must be 'neutral', the others saying that it could 
not be. 

Then Sandford Fleming, the British delegate represent- 
ing Canada, gave an address. He summed up the British 
view: A neutral meridian is excellent in theory but I fear it is 
entirely beyond the domain of practicability.' 1 le then went 
on to read out the following table, showing the number and 
total tonnage of vessels using the several meridians listed 
for finding longitude: 


Initial meridian* 

Ship* of all kind* 

Per call 






37,66 3 
















7 ! 5-448 




















154, [80 










K 1,888 



Rio de Janeiro 

2 53 



' • 





2' • 






'It thus appears that one ol these meridians, that of 
Greenwich, is used by 72 per cent of the whole floating 
commerce of the world, while the remaining 28 per cent is 
divided among ten different initial meridians. If, then, the 
convenience of the greatest number alone should predomi- 
nate, there can be no difficulty in a choice; but Greenwich is 
a national meridian . . .'. Fleming then went on to advocate 


his 1879 proposals that Greenwich plus t8o° should be the 
zero for longitude and time. Passing through the Pacific, he 
said this would have all the advantages of Greenwich but 
could be thought of as being neutral. Fleming's proposal 
was not adopted but the table he quoted had a great 
influence on the final decision. 

After Spain had said she would vote for Greenwich if 
Britain and the USA would adopt the metric system, 
General Strachey announced that Britain had just applied to 
join the international metric convention. However, al- 
though there was no legal bar to its being used in Britain for 
any purpose - and it was already being used extensively by 
scientists - the Government did not expect its use would 
ever be made compulsory. Sir William Thomson (later Lord 
Kelvin), the distinguished British physicist invited to 
address the conference though not a delegate, summed up 
the feelings of many of the delegates: '. . . but it seems to 
me that England is making a sacrifice in not adopting the 
metrical system. ... It cannot be said that one meridian is 
more scientific than another, but it can be said that one 
meridian is more convenient for practical purposes than 
another, and I think that this may be said pre-eminently of 
the meridian of Greenwich. . . .' 

So at last the vote on Resolution II was taken, in the 
afternoon of 13 October 1884. There were 22 ayes, 1 no (San 
Domingo), and 2 abstentions (France and Brazil). The 
43 conference had chosen Greenwich as prime meridian of the 

world - though it must be made clear that this was merely a 
recommendation to the respective governments, not an 
absolute commitment. During the final session, San 
Domingo explained her negative vote. She '. . . was bound 
to regard equity alone on the occurrence of the disagree- 
ment produced by the proposal of the Delegates of France, 
a nation renowned for being one of the first in intellectual 
progress. . . .' She was glad another French proposal was 
accepted almost unanimously [presumably Resolution VII 
about the decimal division of the circle] which was a good 
omen for future unanimous agreement on behalf of the 
general interest of science. '. . . That day will be saluted 
with a cordial hosanna by the Republic of San Domingo, 
which is always ready freely to give its assent to the pro- 
gress of civilisation.' 

43. Airy's transit circle at Greenwich, which has 
defined the Greenwich meridian since 1851; this 
became the prime meridian of the world in 1884. 
From Tlie Illustrated London News, 1 1 Dec. 1880. 

44. Paris Observatory, through which passes 
Ca-—ini's meridian of 1672. 




'That from this meridian longitude shall be 
counted in two directions up lo ISO degrees, east 
longitude being plus and west longitude minus. ' 

Discussion on this resolution - on how to express longi- 
tude - originally proposed by the USA, was longer than 
might have been expected for two reasons. First, it was 
considered fundamental in connection with the reckoning 
of time, as dealt with in subsequent resolutions; secondly, 
the Rome conference had proposed a 360 notation from 
west to east, but there was a strong school of thought that 
considered it should be counted from east to west, while an 
even stronger lobby considered that the current practice of 
counting 180° both ways should not be changed. 

Somewhat inconsequentially, discussion started with the 
presentation by the US General Railway lime Conven- 
tion of a plea that the Meridian Conference should not do 
anything to disturb the standard time system which had 
proved so satisfactory since its introduction the previous 
year. Sweden proposed to count longitude in one direction 
from east to west; Spain from west to east; Great Britain, 
thinking largely of the convenience of navigators, asked, 
'Why change current practice?' Sandford Fleming, how- 
e\ er,differed from his British colleagues, agreeing with 
Sweden and then going on to make a very long speech 
advocating his own Cosmic Time ideas based on the anti- 
meridian of Greenwich (see p. 132 above). Eventually, on 
the proposal of the USA, the resolution quoted above was 
adopted by 14 votes to 5, with 6 abstaining. The way the 
voting went can be seen in Table II (pp. 146-7). 

'That the Conference proposes the adoption of a 
universal day for all purposes for which it mai/ be 
found convenient, and which shall not interior 
with the use of local or other standard time 
where desirable. ' 

The conference now turned from considering longitude 
to considering time, Resolution IV being devoted to general 
principles. Discussion on this was dominated by a long 
speech by W. F. Allen, Secretary of the US Railway Time 

A PRIMF MERIDIAN 1790 - 1884 

Convention, extolling the virtues of the standard-time 
system (which he had personally done so much to promote) 
for use in ordinary life, leaving universal time for science 
and international telegraphs. Replying to those who said 
exact local time was essential, Allen pointed out that, for 
domestic purposes, several countries had, by adopting a 
standard time, already proved this not to be so. Great 
Britain had, said Allen (but see p. 1 14 above), kept Green- 
wich time since 13 January 1848, which differed from local 
mean time by about 8 minutes in the east of the kingdom 
and 22 V2 in the west; Sweden had kept the time of the 
fifteenth meridian east of Greenwich since 1 January 1879 
with east and west difference of 36V2 and 16 minutes respec- 
tively; the United States and Canada had adopted an hour- 
zone system on 18 November 1883 whereby cities like 
Portland (Me.), Atlanta (Ga.), Omaha (Neb.), and Houston 
(Tex.) all now kept a time - without any ill effects or incon- 
venience - which differed twenty minutes or so from local 
mean time. 'Nearly eighty-five percent of the total number 
of cities in the United States of over ten thousand inhabi- 
tants have adopted the new standard time for all purposes, 
and it is used upon ninety-seven and a half per cent of all the 
miles of railway lines.' The resolution was adopted by 23 to 
nil, Germany and San Domingo abstaining. 

1 45 

Thai this universal day is lo be a mean solar 
day; is to begin for all the world at the moment of 
mean midnight of the initial meridian, coin- 
ciding with the beginning of the civil day and 
dale of that meridian; and is to be counted from 
zero up to twenty-four hours. ' 

After a diversion to listen to various letters sent to the 
conference, proposing various prime meridians such as 
Bethlehem, the frontier between Russia and the USA, the 
Great Pyramid, and Le Havre (which is actually on the 
Greenwich meridian), the conference turned from the 
general principles of the proposed universal time to specific- 
definitions, the main discussion centring on one question: 
should the Universal Day start at noon - as was the practice 

o to 
— .£■■§ 

> S 

B Tr.'-C 




> C = 3 TO 

- « q -a 












T3 1 

i ° 
















1 tl c 

f SI'S 

= i/i S3 '? 
■S •_ 2- J 








— . 


■— 1 











Otil ^* Tn ^ ^"* ^» ^* 

V V V 91 V n 

< < 

= Si. & sL £.& 

Z < < < < Z 

E i I 

-3 -e .2 
° £ -5 

.5 .= .= 


< < < 

rj Zj 01 01 Ol r\ Ol TO 



0J CI Ol 01 Ol 01 

.5 = .5 .5 

< < < < 

b 3j b a ^ Ci 
< < < Z < < 

oi to Oi Oi oi r 3 Ol 0! l> ot Oi -_. *_. 
< < 






< ea 


r= O 

2 o 2 


















c c 


yuyo£uuui = 5,j 

U J i U ]l« o 

>, >^ >, >, >,E- >, 
< < < < < Z < 

£. J2 5! §12 ft. o 

< JS<<JS22 
< < 

01 V 01 Ci 're Oi C 

< < < < i. < < 

0) Q Oi re 0> o 

< Z < < JD < Z 

O O - - - *i O 

< < < < z < < 

01 o 01 ll Oi 01 o 

< Z < < < < < 

TO 01 01 01 Ol 

^' < < < < 

J f t & & 

^ XI / < < 

< < 

0i 01 Oi o> 01 

< < < < < 

o o 2 
Z Z 

01 Ol 


O ^ O ^ if 
>, >s >, >, >, 
< < < < < 

Oi oi n oi oi 
< < Z < < 




O !r 
u Oi 

g £ 



I S 


l-l q 1 = I i 1 

TO 3 TO TO O- S 

c_ a: t/; c/> jT) (/> 

2 1 


J) Oj 

a 3 


- O c^ 


M - (N 

S ° 

V) _ *- 

V - - 

«; 2 

TO C D. 

3- °-.2 £ 




> O a x 
< Z 

< Z t5 'o 

I a 

I J 

§ g 

a « 

o- c 

B I 

3 o 

S o. 

8 £-3? 


, 4 8 


of astronomers and, for some nations, at sea - or should 
it conform to civil practice and start at midnight? One of 
the principal uses for the new universal time was for 
astronomy, so it seemed reasonable that astronomers 
should have to make no change in their practice - and make 
no break in their chronologies, which had been established 
in the time of 1 lipparchos - and the scientists at the Rome 
conference had so recommended. On the other hand, the 
other likely users, the operators of the world-wide system 
of electric telegraphs, must start their day at midnight to 
conform to everyday usage. 

Adams (GB), though himself an astronomer, argued 
strongly that astronomers should conform to the civil day 
and that the Universal Day should start at midnight: noon 
was, after all, mid-day, or midi, not the beginning or the end 
of the day. The advantages of retaining astronomical tradi- 
tion on this point, he said, seemed small compared with the 
overwhelming disadvantages which would be imposed by 
having two conflicting methods of reckoning dates. 'If this 
diversity is to disappear, it is plain that it is the astronomers 
who will have to yield. They are few in number compared 
with the rest of the world. They are intelligent, and could 
make the required change without any difficulty, and with 
slight or no inconvenience . . .' -a sentiment not shared by 
the majority of his astronomical colleagues, as we shall 

A Swedish resolution, supported by Austria-Hungary, 
Italy, the Netherlands, Switzerland, and Turkey, in favour 
of starting the day at noon, was lost by 6 to 14 with France, 
Germany, San Domingo, and Spain abstaining. The 
Turkish delegate pointed out that, whatever the conference 
might recommend, in Turkey there would always be two 
times kept, I'heure a la franque from midnight to midnight, 
and I'heure a la turque from sundown to sundown, used by 
agricultural workers, from which Muslim prayers were 

The resolution quoted above, for the Universal Day to 
start at midnight on the prime meridian, was eventually 
passed by 15 to 2 - Austria-Hungary and Spain - with 7 
abstentions, reversing the proposals of the Rome 

A PRIME MERIDIAN 1790 - 1884 

'That the Conference expresses the hope that as 
soon OS may he practicable the astronomical ami 
nautical day* will he arranged everywhere to 
begin at mean midnight. ' 

We have already seen how astronomers (following 
Hipparchos) used a 24-hour-clock system and (until 1925) 
began their day with the Sun's culmination at noon: this was 
convenient because it meant that the date did not change in 
the middle of a night's observing, as it would if civil reckon- 
ing had been used. At sea until the nineteenth century, on 
the other hand, the day was considered to end at noon on 
the civil day concerned. The log was kept in ship's time, that 
is local apparent time, adjusted at intervals for change 
of longitude: and by old practice the day would end (and 
the time be adjusted) at noon when the officer taking the 
latitude sight called 'Twelve o'clock, sir', and the Captain 
would reply 'Make it so!' Thus, noon marked the begin- 
ning of the day - say Monday - in astronomical reckon- 
ing, the middle ('midday') in civil reckoning, and the end of 
the day in nautical reckoning. While 6 a.m. Monday civil 
reckoning was also 6 a.m. Monday to the navigator, it was 
18.00 Sunday to the astronomer. On the other hand, twelve 
hours later, at 6 p.m. Monday civil reckoning, it was 6 p.m. 
Tuesday to the navigator but 6.00 Monday to the 
astronomer. All of which was further confused because. 


Civil reckoning 
Day ends a! midnight 

Midnight Noon Midnight Noon Midnight Noon Midnight 

a.m. : p.m. 

a.m. p.m. 

12 6 12 6 1; 

a.m. p.m. 

12 6 12 6 12 

Astronomical S 

reckoning '-SATURDAY 

Day begins at noon < 




12 IB 00 OS 12 18 00 06 12 18 00 06 12 

Nautical reckoning (aiauncv 
Day ends a, noon >JEJ™ 

p.m. i a.m. 

p.m. I a.m. 


p.m. |a.m.^ 

12 6 12 6 12 6 12 6 12 6 1? S 12 
Midnt Noon Midnt Noon Midnt Noon Midnt 

45- The civil,, ,ind nautical daws. 



when he entered harbour, the navigator (and the ship's log) 
reverted to civil reckoning. None of this mattered greatly 
until the Nautical Almanac (which followed astronomical 
practice) began to be used at sea, after which the opportuni- 
ties for errors began to multiply. For example, the historian 
of, say, Cook's voyages has to be careful today because the 
journals of Cook himself and of his astronomer William 
Wales each used a different reckoning for recording the 
same events, and these differed again at sea and in 

harbour. 36 

At the time of the Washington conference the Astro- 
nomical Day was very much alive but the Nautical Day had 
been abandoned in favour of civil reckoning by seamen of 
many nations. In the Royal Navy the demise of the Nautical 
Day occurred with the issue of an Admiralty Instruction 
dated 1 1 October 1805 (ten days before the Battle of 
Trafalgar) describing a new form of log-book 'to be kept in 
all King's ships. . . calendar or civil day is to be made use of, 
beginning at midnight. . . . It is necessary to remark to you, 
that the private night signal for each day of the month is to 
continue in force until clay-light of the following day' 37 - 
which last observation was, of course, for the same reason 
that astronomers preferred a day beginning at noon. Apart 
from the King's ships, the Nautical Day was abandoned by 
the East India Company in the 1820s but remained in use in 
many other merchant ships until well into the middle of the 
nineteenth century. As was made evident at the conference, 
however, the Nautical Day was still in use in ships of some 
nations at that time. 

However, there had been so much debate on the matter of 
the time at which the Universal Day should start for the 
previous resolution that Resolution VI was carried with 
almost no discussion and without a division. Nevertheless, 
despite the apparent unanimity, it was to be thirty-five 
years before the recommendation was realized. 

'That the Conference expresses the hope that the 
technical studies designed to regulate and extend 
the application of the decimal system to the divi- 
sion of angular space and of time shall be re- 

it PRIME MERIDIAN 1790- 1884 

sumed, so as to permit the extension of this 
application to all cases in which it presents real 


This resolution, following a similar one in the Rome con- 
ference, was proposed by France. After some discussion as 
to whether the subject of decimalization of angles and time 
was not outside the conference's terms of reference, it was 
put to the vote and adopted by 21 to o, with Germany, 
Guatemala, and Sweden abstaining. 

Before the conference ended, Great Britain proposed two 
more resolutions, one recommending standard time for 
local civil time 'at successive meridians distributed around 
the earth, at time-intervals of either ten minutes, or some 
integral multiple of ten minutes, from the prime meridian'. 
(The Swedish astronomer Gyldcn had proposed a standard 
time system based on time intervals of 2'/;°, or 10 minutes of 
time.) The other resolution proposed that arrangements for 
the adoption of the Universal Day in international tele- 
graphy should be left to the International Telegraph Con- 
gress. However, it was decided that the questions raised 
were covered by resolutions already adopted, so these last 
two were withdrawn. 

And there, except for a summing up in the Final Act and 
some short speeches of thanks on 1 November, ended the 
International Meridian Conference of 1884, recommending 
(among other things) that the Universal Time should be 
Greenwich Time. It is interesting that the three resolutions 
recommending matters of principle - the desirability of 
having a single prime meridian, a Universal Day, and 
decimal angle and time -were carried almost unanimously. 
Of the three specific resolutions defining the prime meridian 
and universal time, Great Britain and the USA voted in 
favour, and were followed by a substantial number of 
countries. On the other side, Brazil, France, and San 
Domingo abstained or voted against. Austria-Hungary, 
Germany, Italy, the Netherlands, Spain, Sweden, Switzer- 
land, and Turkey then supported the UK/USA bloc on the 
choice of the Greenwich meridian, but abstained or voted 
against on the other issues. 

O Greenwich time for the world 1884-1939 

Standard time 

The principal impact of the Washington conference on the 
man-in-the-street was the adoption, country by country, of 
a time-zone system based upon the world's new prime 
meridian, Greenwich - and this despite the fact that the 
time-zone system, though discussed, was not specifically 
recommended by the conference. 

At first, many of those who conceived the Universal Day 
and Universal Time - or Terrestrial Time, or Cosmic Time 
- envisaged one single time being used all over the world 
for all purposes. But this view never prevailed beyond 
one or two enthusiasts. However, Dowd's time-zone 
system with one-hour differences between zones seemed to 
many to be the best possible compromise between Uni- 
versal Time and local time: no one had to keep his clock 
much more than 30 minutes different from local time, and 
yet the minute hands were all the same, only the hour 
hands differing from zone to zone. To those who com- 
plained that the Sun was no longer precisely on the 
meridian at noon, it was pointed out that the change from 
Apparent Time to Mean Time - by then a fait accompli in 
civilized countries the world over - was an even more 
fundamental change than that from local mean time to 
Standard Time. How many people realize that, as a result 
of the change from apparent to mean time, afternoons in 
November are almost half an hour shorter than the morn- 
ings, while in February the mornings are half an hour 
shorter than the afternoons? 

While the conference was meeting, four countries were 
a I ready keeping the new system - Great Britain, Sweden, 
the USA, and Canada. As can be seen in Table III, a 


standard time based on the Greenwich meridian was adop- 
ted country by country, often first for railway and telegraph 
purposes, followed soon for legal and general purposes. 
By 1905 the only major countries not conforming were 
France, Portugal, I lolland, Greece, Turkey, Russia, Ireland 
(oddly enough), and most of Central and South America 
except Chile. Of thirty-six nations then using Standard 
Time, twenty had adopted Greenwich as the basis of their 
system; of the remaining sixteen, no two agreed. France did 
not conform at first but, with Algeria, adopted Paris Mean 
Time as 1'lwuiv uationak' by the law of 14 March 1891 . As had 
happened in many other countries, this was largely a legal- 
ization of the railway time already being generally used 
throughout the country. In December 1891 E. Pasquier 
wrote in Cicl ct Tare that 'the almost unanimous adhesion of 
civilized nations to the meridian of Greenwich should cause 
the partisans of other meridians to lay down their weapons, 
and henceforth all efforts should be directed to settling 
definitely an hourly unification at once simple, rational and 
practical.' 1 

Some five years later, on 27 October 1896, Deputy Deville 
introduced a Bill into the Chamber of Deputies proposing 
Greenwich Mean Time as the legal time in France. The Bill, 
with the amendment that this should be expressed as Paris 
Mean lime diminished by 9 minutes 21 seconds (which 
comes to the same thing as GMT but avoids the use of the 
word Greenwich), was passed on 24 February 1898 by the 
Chamber to the Senate, who duly passed it to a Commission 
- and there it stayed for twelve long years. Apparently the 
Ministries of Commerce, Industry, Posts and Telegraphs, 
and Public Works all accepted the Bill but it was strongly 
opposed by the Ministries of Public Instruction and the 

Eventually, on 9 March 1911 (by which time radio had 
become a reality), the following law was passed and came 
into effect on the night of 10-1 1 March: 'Legal time in France 
and Algeria is Paris Mean Time, retarded by 9 minutes 21 
seconds.' 2 This law remained in force until 9 August 1978 
when it was repealed by a decree which stated that hence- 
forward French legal time should be determined from Co- 
ordinated Universal Time (UTC, which by definition is kept 



Table III Dates of adoption of zone times based on the 
Greenwich meridian, including half-hour zones 
N.B. (1) The names of countries given here are those in general use 
on the date quoted. 

(2) In some cases, standard times were brought into use for 
railways and telegraphs before the dates stated. 

(3) Countries with no sea coast are, in general, omitted. 
1848 Great Britain (legal in 1880) 

1879 Sweden 

1883 Canada, USA (legal in 1918) 

1884 Serbia 
1888 Japan 

1892 Belgium, Holland*, S. Africa except Natal 

1893 Italy, Germany, Austria-I Iungary (railways) 

1894 Bulgaria, Denmark, Norway, Switzerland, Romania, Turkey 

1895 Australia, New Zealand, Natal 

1896 Formosa 

1899 Puerto Rico, Philippines 

1900 Sweden, ligypt, Alaska 

1901 Spain 

1902 Mozambique, Rhodesia 

1903 Ts'intao, Tientsin 

1904 China Coast, Korea, Manchuria, N. Borneo 

1905 Chile 

1906 India (except Calcutta), Ceylon, Seychelles 

1907 Mauritius, Chagos 

1908 Faroe Is., Iceland 

1911 France, Algeria, Tunis and many French overseas pos- 
sessions, British West Indies 

1912 Portugal and overseas possessions, other French pos- 
sessions, Samoa, Hawaii, Midway and Guam, Timor, Bismarck 
Arch., Jamaica, Bahamas Is. 

1913 British Honduras, Dahomey 

1914 Albania, Brazil, Colombia 

1916 Greece, Ireland, Poland, Turkey 

1917 Iraq, Palestine 

1918 Guatemala, Panama, Gambia, Gold Coast 

1919 Latvia, Nigeria 

1920 Argentine, Uruguay, Burma, Siam 

1921 Finland, Estonia, Costa Rica 

1922 Mexico 

•Legal time reverted to Amsterdam time 1909; to Central European Time 


1924 Java, USSR 

1925 Cuba 

1928 China Island 

1930 Bermuda 

1931 Paraguay 

1932 Barbados, Bolivia, Dutch East Indies 
1934 Nicaragua, E. Niger 

By 1936 I-abrador, Norfolk I. 

By 1937 Cayman Is., Curacao, Ecuador, Newfoundland 

By 1939 Fernando Po, Persia 

1940 I lolland 

By 1940 Lord I lowe I. 

By 1948 Aden, Ascension I., Bahrein, British Somaliland, 
Calcutta, Dutch Guiana, Kenya, Federated Malay States, Oman, 
Straits Settlements, St. I lelena, Uganda, Zanzibar 

By 1953 Raratonga, South Georgia 

By 1954 Cook Is. 

By 1959 Maldive I. Republic 

By 1961 Friendly Is., Tonga Is. 

By 1962 Saudi Arabia 

By 1964 Niue Is. 

1972 Liberia 

In 1978, Guyana was keeping + 3h 45m; Chatham Island — 12b. 
45m. Otherwise, all countries were keeping time within an even 
hour or half-hour of Greenwich. 


Principal sources: 

Koppenstiitter (ed.), Zonen und Sommerzeiten alter Under und Sifidte 

den f>(/c(Munchen I1937I). 
US National Bureau of Standards. 'Standard Time throughout the 

World', Circular of the Bureau of Standards, no. 399(i5Sept. 1932). 
The Observatory, Feb. 1901, 88-91. 
Abridged Nautical Almanac annually. 


within 0.9 seconds of GMT), legal time being obtained by 
adding or subtracting an exact number of hours to UTC. The 
significance of this decree is discussed in Chapter 7. 3 

Of the other countries mentioned above, Portugal 
adopted a time based on the Greenwich meridian in 1912, 
Brazil and Colombia in 1914, Greece, Ireland, Poland, and 
Turkey in 1916, Argentina and Uruguay in 1920, the USSR 
in 1924. Holland, whose railways had kept GMT since 1892 
but who had used Amsterdam mean time instead of local 
time for other purposes since 1909, was forced to adopt 
Central F.uropean Time on 19 May 1940 during the German 
46 Occupation, a usage which was finally confirmed in 1956. 

The last major nation to conform was Liberia, whose legal 
time remained 44m 30s slow on GMT until January 1972 
when, in honour of the President's birthday, she adopted 

Unification of civil and astronomical days 

Astronomers, however, were more conservative. Despite 
Resolution VI of the Washington conference being passed 
without a division, there was very considerable opposition 
from certain parts of the astronomical world to the idea of 
the astronomical day starting at midnight instead of noon. 
This was voiced volubly at an astronomical congress in 
Geneva the following year when the Washington Resolu- 
tion was condemned by the majority of the astronomers 
present, particularly Newcomb (USA), Auwers (Berlin), 
Gylden (Stockholm), and Tietjen (Berlin), though defended 
by O. Struve (Pulkowa). In passing, it is worth noting that, 
as long ago as 1804, Laplace had proposed this unification 
and, after a long discussion, his view had been adopted by 
the Bureau des Longitudes by 7 votes to 5, though the 
Coiinaissancc des Temps remained faithful to the old 
notation. 4 

In 1885 Britain adopted the Civil Day for spectroscopic, 
photographic, magnetic, and meteorological observations, 
while the 24-hour clocks in Greenwich Observatory (in- 
cluding the public clock outside the gates)were, on 1 Januarv 
1885, set to civil time, with the day starting at midnight. 
However, the astronomical observations and the Nautical 
Almanac remained in astronomical time, pending general 


international agreement. In 1893 tne J omt Committee of the 
Canadian Institute and the Astronomical and Physical 
Society of Toronto, under the chairmanship of Sand ford 
Fleming, sent a circular to astronomers of all nations, asking 
the question: 'Is it desirable, all interests considered, that on 
and after the first day of Januarv, 1901, the astronomical day 
should everywhere begin at Mean Midnight?' 5 Of the 171 
replies received, 108 were in favour, 63 against. Broken 
down into nationalities, 18 countries were in favour (includ- 
ing the United States, though Simon N'ewcomb did not 
reply), 4 against (Germany, I lolland, Norway, Portugal). 

About the same time W. M. Greenwood of Glasson Dock, 
Lancaster, England, sent a circular to shipmasters of all 
nations asking four questions broadly covering the main 
recommendations of the Washington conference. Of 409 
replies received to Question 3 on the matter of midnight t>, 
noon, 399 chose the former, 10 the latter. On the 24-hour- 
clock system, there were 22 against.'' In France the Bureau 
des Longitudes, when asked by the Ministry of Foreign 
Affairs in 1884 to give an opinion, voted 7 to 5 in favour of 
the change. 7 But, despite these opinions, no general agree- 
ment was reached among astronomers and the nautical 
almanacs continued to use noon to begin their day. Then in 
1917, at the Anglo-French Conference on Time-keeping at 
Sea (mentioned below), the following resolution was 

1 1 . That, from the point of view of seamen it would be a consider- 
able advantage if a day commencing at o hours midnight were 
substituted for the astronomical day commencing at o hours noon 
in all nautical publications, and that the Royal Astronomical 
Society should be asked to ascertain the views of astronomers as to 
such an alteration, including possibly the general substitution of 
the civil for the astronomical day. 8 

Although there were dissenting views, the replies to the 
RAS's circular were generally favourable to the change, 
many astronomers grudgingly agreeing that, if it would 
help seamen, it should be done. So, from the issue for 1925, 
the British and many other nautical almanacs began to use 
the civil day, beginning at midnight. 

However, the arguments continued among astronomers, 
largely centred on nomenclature: before 1925, the GMT 


46. Time zone Chart, 1979. Zones are 15°( = 1 hour) wide. The /.one number 
indicates the correction in hours to be applied to zone lime to obtain GMT 
(or Universal Time); e.g. I4.oo(- 10) in Sydney. Australia, is the equivalent 
of 04.00 GMT. /tone letters (Z = GMT, K = -10, for example) are used to 

indicate the zone time being used in international communications. Based 
on Admiralty chart A3, produced with the sanction of the Controller. I IN! 
Stationery Office, and of the I lydrographer of the Navy. 


day began at noon; should a time-scale differing 12 hours 
from this still be called by the same name? In America, 
Greenwich Civil Time (GCT) was favoured as the name for 
the new time scale; the British Admiralty disagreed because 
of the possibility of confusion with a warship's Gunnery 
Control Tower - GCT. Then, in 1928, the International 
Astronomical Union recommended internationally that the 
time scale used in almanacs should be called Universal Time 
(UT). Thereafter, UT began to be adopted for astronomy 
though the term GMT is still used in navigational publica- 
tions, in rail and air timetables, and for international cable 
and radio communications. 9 

The International Date Line 

A corollary to the establishment of a prime meridian for 
longitude and time is that there should be an anti-prime 
meridian i8o°away-an International Date Line, on either 
side of which the date is different. At one particular instant, 
it will be Monday to the west of that line but Sunday to the 
east of it. The problem of this change of date came to light 
as early as the first circumnavigation. When the sur- 
vivors of Ferdinand Magellan's expedition reached civiliza- 
tion in 1522, having sailed through the Straits now called 
after him and then westwards across the Pacific to the 
Philippines (where Magellan himself was killed) and the 
Spice Islands, and then around the Cape of Good Hope, 
they discovered, not only that they had practical proof that 
the Earth was round, but also that they had somehow 
gained a day in their lives, as was explained by Antonio 
Pigafetta, the young Italian nobleman from whose diary we 
learn so much about the voyage. Despite the fact that the 
Cape Verde Islands were Portuguese and therefore enemy 
territory, the Spaniards in the one remaining ship, the 
Victoria, were forced to call there for supplies. According to 
Pigafetta's diary, it was Wednesday 9 July 1522, but, 

In order to see whether we had kept an exact account of the days, 
we charged those who went ashore to ask what day of the week it 
was, and they were told by the Portuguese inhabitants of the island 
that it was Thursday, which was a great cause of wondering to us, 
since with us it was only Wednesday. We could not persuade 
ourselves that we were mistaken; and I was more surprised than 


the others, since having always been in good health, I had every 
dav, without intermission, written down the day thai was current. 
But we were afterwards advised that there was no error on our 
part, since as we had always sailed towards the west, following the 
course of the sun, and had returned to the same place, we must 
have gained twenty-four hours, as is clear to any one who reflects 
upon it. 1 " 

More than 150 years later, the English circumnavigator 
William Dampier (1652-1715) came across another aspect of 
the problem. Having travelled westwards around the world 
(as Magellan had done), Dampier's ship reached the 
Philippines on 14 January 1687: 

It was during our stay at Mindanao, that we were first made sensible 
of the change of time, in the course of our Voyage. For having 
travell'd so far Westward, keeping the same Course with the Sun, 
we must consequently have gain'd something insensibly in the 
length of the particular Days, but have lost in the tale, the bulk, or 
number of the Days or I lours. According to the different Longi- 
tudes of England and Mindanao, this Isle being West from the 
Liztard, by common Computation, about 210 Degrees, the differ- 
ence ol time at our Arrival at Mindanao ought to be about 14 Hours: 
And so much we should have anticipated our reckoning, having 
gained it by bearing the Sun company. Now the natural Day in 
every particular place must be consonant to itself: But this going 
about with, or against the Sun's course, will of necessity make a 
difference in the Calculation of the civil Day between any two 
places. Accordingly, at Mindanao, and all other places in the Easl- 
ludies, we found them reckoning a Day before us, both Natives and 
European*; for the E ur opea n s coming eastward by the Cape of Good 
Hope, in a Course contrary to the Sun and us, where-ever we met 
they were a full Day before us in their Accounts. So among the 
Indian Mahometans here, their Friday, the Day of their Sultan's 
going to their Mosques, was Thursday with us; though it were 
Friday also with those who came eastward from Europe. Yet at the 
Ladrone Islands, we found the Spaniards of Guam keeping the same 
Computation with ourselves; the reason of which I take to be, that 
they settled that Colony by a Course westward from Spain; the 
Spaniard* going first to America, and thence to the Ladrone? and 
Philippines. But how the reckoning was at Manila, and the rest of the 
Spanish Colonies in the Philippine Islands, I know not; whether they 
keep it as they brought it, or corrected it by the Accounts of the 
Natives, and of the Portugiteze. Dutch and English, coming the 
contrary way from Europe. ' ' 




Thus there was a discrepancy as to date around the borders 
of the Pacific according to whether the colonizers came from 
the east or the west. The Portuguese, then the Dutch, the 
French, and the British, came to the East Indies by way of 
the Cape of Good Hope. The Spaniards, on the other hand, 
reached the Philippines and Ladrones from America. Until 
1844 the Philippines kept 'American date' while Celebes, in 
the same longitude, kept 'Asiatic date'. 

To the north there was another example of 'reaching into 
the wrong hemisphere', this time with Asia reaching into 
America. Inspired by Bering's discoveries, Russian fur- 
traders settled in Alaska as early as 1745. In course of time 
Alaska became a Russian colony, in which the Orthodox 
Church, with its Julian calendar, was established, the dates 
being the same as those in St. Petersburg and Moscow. In 
the nineteenth century American traders streamed into 
Alaska causing worry to the Orthodox priests, not only by 
their adherence to the Gregorian calendar (which affected 
the day of the month) but also because they insisted upon 
observing their day of rest on the day that the Russians 
claimed to be a Monday. Eventually, in 1867, the Territory 
was bought by the United States for seven million dollars, 
and the Gregorian calendarand American date introduced. 

In 1879 the British governor of the Fiji Islands (through 
which the 180° meridian passes) enacted an ordinance to say 
that all the islands should keep the same time, Antipodean 
Time. The King of Samoa, however, under pressure from 
American business interests, decided to go the other way, 
changing the date in his kingdom from the Antipodean to 
the American system, ordaining - by a masterpiece of 
diplomatic (lattery - that the Fourth of July should be 
celebrated twice in that year. '- 

The date line as originally drawn had a kink to the west- 
ward of the I Iawaiian Islands to include Morrell and Bvers 
islands which appeared on nineteenth-century charts at the 
western end of the I Iawaiian chain. It was then proved that 
they did not exist, so the date line was straightened out. 
The Cook Islands remain on a different side of the date line 
from New Zealand, by whom they are administered. There 
is a saying in Raratonga: 'When it's today in Raratonga, it's 
tomorrow in Wellington.' 


It is worth noting that the drawing of the International 
Date Line is not the result of a formal international agree- 
ment but, in the words of the Hydrographer of the Navy in 
Britain, is 'merely a method of expressing graphically . . . 
the differences of date which exist among some of the island 
groups in the Pacific'. 

Radio time signals 

The radio time signal was a fundamental step in the de- 
velopment of the dissemination of time, particularly for 
navigational purposes. At last it was possible for a ship to 
check her chronometers out of sight of land. The radio time 
signal also drove the final nails into the coffin of the lunar- 
distance method of finding longitude at sea. With chron- 
ometers cheap enough to be carried in ships of every size, 
lunar distances had long been superseded for day-to-day 
use, though the ability to use them occasionally had never- 
theless to be retained so that the navigator could check his 
chronometers when other means were not available. Radio 
removed the need for this and the British Nautical Almanac 
ceased to publish lunar-distance tables in 1907, though in- 
structions on how to compute and reduce them continued 
until 1924. One of the most important dates in our story was 
therefore 29 March 1899 when Guglielmo Marconi, from 
near Boulogne with an apparatus invented by the French- 
man Edouard Branly, detected a signal sent over the English 
Channel from near Dover (all of 45 km distant). So wireless 
telegraphy, soon to be called radio-telegraphy or just plain 
radio, proved to be a practical means of communication, a 
proof reinforced when, two years later, Marconi was in 
Newfoundland to receive the first transatlantic radio 

The earliest wireless time signals for navigational pur- 
poses seem to have been those broadcast by the US Navy on 
low power from Navesink, NJ, in the spring of 1904, leading 
to the first regular transmissions every day at noon EST 
from Washington, DC, from January 1905. High-power 
radio time signals from Arlington, Va., began in December 
1912. ,3 In Germany, experimental transmissions began 
from Norddeich Radio (30 km north of Emden) in 1907, 
regular transmissions from May 1910. 





France had never had a full-scale time service controlled 
by a national observatory as Britain had had since 1852, and 
it was not until 1880 that telegraphic time signals were sent 
to towns in France that wanted them, and then only once a 
week on Sundays, at first to Rouen and I e I la\ re, later to I .a 
Rochelle, Nancy, Saint-Nazaire, Chamber)', and Cluses. M 
In 1908, however, the Bureau des longitudes rec- 
ommended that wireless time signals should be broadcast 
regularly from the Eiffel Tower, a project which the Ministry 
of War agreed to sponsor. All was ready by January 1910 
but, just as the service was about to start, there was a 
particularly bad flooding of the Seine and underground 
installations on the Champ-de-Mars were inundated. 
Repairs were made and regular transmission of time signals 
started on 23 May 1910, daily at midnight Paris Mean Time. 
On 21 November a daytime signal commenced at 11.00 
daily. On 9 March 191 1, by the law of the same date already 
mentioned, France put her clocks back to the equivalent of 
GMT but the Eiffel Tower did not follow suit until 1 July 
when time signals started at 10.45 ar, d 2 3-45 GMT daily. 
From February 1912 an additional rhythmic signal on the 
vernier principle was broadcast, allowing the error of a 
clock or watch to be found to an accuracy of a hundredth of 
a second. 15 Similar 'scientific' time signals were broadcast 
from many other countries after the Second World War; that 
from Rugby, England, from 1927 to 1958. 

There were no early moves in England to institute wire- 
less time signals, which would, of course, have been of 
particular value to shipping. This is surprising but perhaps 
it was thought that, in time of peace, the time-balls in 
various ports - and listening to foreign time signals - would 
he adequate, while, in time of war, radio time signals would 
be stopped anyway. Nevertheless, a wireless room was 
fitted up at Greenwich Observatory so that foreign signals 
could be received, compared with the time determined at 
Greenwich, and discrepancies reported to the observatory 

The Bureau International de l'Heure 

By 191 1 it had been found that wireless time signals sent 
from the various stations could differ from each other by 


several seconds. Having removed one of the obstacles to 
international co-ordination in this matter by adopting 
GMT, the French took the initiative and, in May 1912, 
invited certain other governments to send delegates 'to 
studv ways and means of effecting practical unification of 
radio time signals, and to prepare plans for an international 
time service to suit the needs of all'. "' 

The conference, with representatives from sixteen states, 
assembled at Paris Observatory on 15 October 1912, the 
directors of most national observatories being present: 
Britain sent her new Astronomer Royal, F. W. Dyson, and 
the Assistant Hydrographer, Capt. J. F. Parry; the USA sent 
Prof. Asaph Mall of the Naval Observatory; Germany sent 
Prof. W. Foerster who had been director of Berlin Obser- 
vatory. After lengthy discussions, a formal proposal was 
made to establish an International Time Commission which 
would have three main objectives: to secure the unification 
of time signals; to secure the universal use of GMT; and to 
create an international organization to be called the Bureau 
International de l'Heure (BIH), with the task of co-ordi- 
nating results from observatories and deducing the most 
exact time - I'heure definitive. An interesting world-wide 
network of time signals was also proposed, to start on 1 July 
1913, a target date which was to be met by few of the 
countries concerned: 




Fernando I. (Brazil) 

Arlington (USA) 

Manila (Philippines) 

Mogadishu (Somalia) 




Fernando I. (Brazil) 

Arlington (USA) 

Massawa (Eritrea) 

San Francisco (USA) 









A second International Time Conference met in Paris in 


October of the following year. This was of a diplomatic 
rather than a scientific character, was attended by repre- 
sentatives of thirty-two states, and led to the drawing-up of 
statutes for a new body to be called the International Time 
Association whose principal purpose was to superintend 
the proposed BIH. At the request of a committee appointed 
by the conference, the director of Paris Observatory set up a 
provisional Time Bureau in 1913 in anticipation of the ratifi- 
cation by the various states of the conference proposals. In 
the event, the outbreak of the First World War less than a 
year later prevented full ratification. Despite many diffi- 
culties, however, the provisional BIH continued to operate 
throughout the war. 

In 1918 the Royal Society in London took the initiative in a 
move to re-start the international scientific co-operation 
which had been developing before the war by calling the 
Inter-Allies Conference of Scientific Academies, which met 
in London in October 1918, in Paris in .November 1918, and 
in Brussels in July 1919. At this last meeting, the Inter- 
national Astronomical Union (1AU) was formed, meeting 
the same year and immediately setting up a Time Com- 
mission within its own structure to serve the same purpose 
as the proposed lime Association of 1913, principally to 
supervise the BIH which was finally established on an inter- 
national basis on 1 January 1920, operating from Paris 
Observatory, with Guillaume Bigourdan (1851-1932, who 
had given the opening address at the i9i2conference)asits 
first director. 17 

Daylight Saving Time 

Daylight Saving Time, or Summer Time as it is generally 
known in Britain, was the brain-child of William Willett 
(1857-1915), a London builder living at Petts Wood in Kent. 
In a pamphlet circulated in 1907 to many Members of Parlia- 
ment, town councils, businesses and other organizations, 
he pointed out 'that for nearly half the year the sun shines 
upon the land for several hours each day while we are 
asleep, and is rapidly nearing the horizon, having already 
passed its western limit, when we reach home after the 
work of the day is over . . .'.' 8 He proposed to improve 
health and happiness by advancing the clocks 20 minutes 


on each of four Sundays in April, and retarding them by the 
same amount on four Sundays in September. In addition to 
improvements in health and happiness - his main objec- 
tive - Willett also claimed that, with electricity costing Vioof 
a penny per hour, the country would save £2'/2 million, 
even taking into account the loss of profit to producers of 
artificial light. Though the scheme was ridiculed and met 
with considerable opposition, particularly from farming 
interests in England, nevertheless a Daylight Saving Bill 
was drafted in 1909 and introduced in Parliament several 
times, though it met with no success before war broke out. 
In April 1916, however, Daylight Saving Time was in- 
troduced as a wartime measure of economy, not only in 
Britain but, within a week or so, in nearly all countries, 
both allied and enemy. Willett had died the previous year so 
never saw his ideas put into effect. Though many countries 
abandoned Daylight Saving Time immediately after the 
war, most reintroduced it eventually, and some even began 
to keep it throughout the year. In the Second World War 
Britain kept 'Summer Time' (BST) in the winter, with 
'Double Summer Time' (DBST, two hours in advance of 
GMT) in the summer. 

Between 1968 and 1971 Britain tried the experiment ot 
keeping BST- to be called British Standard Time - through- 
out the year, largely for commercial reasons because Britain 
would then conform to the time kept by the rest of the 
countries in the European Community. This measure met 
with considerable opposition from the country generally - 
and the more so the further west and north. In the summer, 
no one minded; in the winter, however, the children of 
Glasgow and points north, for example, always had to go to 
school in the dark. The experiment was abandoned and, 
since 1972, Britain has kept GMT in winter and BST in 
summer. Most other countries now keep Daylight Saving 
Time in summer, some (like France) throughout the year. 
'Spring forward, fall back' - so runs the only mnemonic the 
author knows for deciding that problem: 'which way should 
the clocks go?' 

Standard time at sea 

As we have seen, the standard time system based on the 


47- Wireless time signals .it sea, iqio. 

48. Wireless lime signals in Britain in the home, 1923. Loudspeaker reception with Marconiphone 


Greenwich meridian (sometimes called the zone-time 
system) quickly established itself ashore. At sea, however, 
there was at first no such agreement. Though the navigator 
used GMT for his calculations, for domestic purposes afloat 
it was the practice to put the clocks forward or back so as to 
make them show the exact apparent time of the vessel's 
noon position. In wartime, this was extraordinarily in- 
convenient. In a convoy, for example, the Commodore 
could not simply say. 'The convoy will alter course at noon.' 
He had to specify the longitude he was using - because 
every ship in the convoy was keeping a different noon. 

In June 1917, therefore, an Anglo-French Conference 
on Time-keeping at Sea assembled in London. This rec- 
ommended that the zone-time system should be used at 
sea, clock changes required by changes of longitude being 
made preferably in one-hour steps. This recommendation 
was immediately adopted by British and French ships, both 
naval and mercantile. Ships of most other nations soon 
followed suit so that, by a few years after the war, zone time 
was kept at sea, certainly by almost all naval ships and by 
many non-naval ships as well. Nevertheless, up to the 
Second World War, the old practice of changing ship's time 
at midday prevailed in many independent merchant ships. 

Greenwich time in the home 

As far as domestic time signals in Britain are concerned, 
GMT seems first to have come into the home by radio when 
the British Broadcasting Company (before it became a cor- 
poration) broadcast the chimes of Big Ben to usher in the 
year 1924. Late in 1923 Frank Dyson, the Astronomer Royal, 
had visited John Reith, Director General of the BBC, and 
discussed the idea of public time signals being broadcast. 
The famous 'six-pip' time signal - 'pips' to mark seconds 
55' 56, 57. 58, 59, 60 - was Dyson's brain-child, devised in 
discussion with Frank I lope-Jones, inventor of the free- 
pendulum clock, who had himself originally advocated a 
'five-pip' signal. 

On 5 February 1924 Dyson broadcast to the nation, 
inaugurating the new service. A little later he was presid- 
ing over a dinner of the British I lorological Institute when 
Hope-Jones was guest of honour. Some wag, remembering 




the latter's connection with time signals, handed him six 
orange pips on a plate. With much ceremony, Hope-Jones 
formally presented the sixth pip to Dyson in the chair! 1 '' 

The speaking clock 

Time by telephone has a fairly long history. In 1905 Paris 
Observatory established a telephonic time signal available 
on demand, by placing a microphone in the mean-time 
clock case when needed, transmitting the clock ticks while 
at the same time a member of the observatory staff counted 
the minutes and seconds verbally on another instrument on 
the same line. In 1909 Hamburg Observatory established a 
slightly less precise telephone time service, also on de- 
mand. 20 The original Paris telephone time service proved 
very popular - but very time-consuming for observatory 
staff. On 14 February 1933, therefore, a new service - 
I'horloge parlante - was offered by Paris Observatory. This 
was a completely automatic 'speaking clock', available to 
any telephone subscriber asking for, or dialling, the appro- 
priate number. Similar systems were already in use in 
Strasbourg and outside France. 21 

A similar 'speaking clock' was brought into use in Britain 
on 24 July 1936. On dialling TIM the subscriber would hear 
the time every ten seconds from the 'golden voice': 'At the 
third stroke, it will be six, fifty-seven, and twenty seconds.' 

The meridian moved 

The official maps of Great Britain are published by a military 
organization known as the Ordnance Survey, the original 
triangulation for which took place between 1783 and 1853. 
Between 1938 and the 1950s a retriangulation of Great 
Britain was carried out. As part of this, observations were 
taken in 1949 to connect the other stations of the survey with 
Airy's transit circle, which, by definition, must be oo°oo'oo" 
in longitude. In fact it transpired that, even after the angles 
had been reobserved, the retriangulation gave a longitude 
value for Airy's transit circle of oo°oo'oo".4i7 east of 
Greenwich - which should be impossible. Consternation! 
This discrepancy in longitude which was the equivalent of 
8.04 metres (or 26.39 feet) on the ground, was far greater 
than might be expected from errors of observation, even in 


the eighteenth century. There was a discrepancy in latitude 
as well, amounting to 0.039 seconds of arc (=3.95 ft = 1.21 
m), but this was acceptable. 

The person who provided the answer to the puzzle was 
Ihe Royal Observatory's then Chief Assistant, Dr. R. d'E. 
Atkinson. He pointed out that, when General Roy had 
carried out the Principal Triangulation in 1787, the great 
theodolite had been erected immediately over Bradley's 
transit instrument which then defined the Greenwich 
meridian. Pond's transit instrument replaced Bradley's in 
1816, but it was mounted on the same piers, so there was no 
change in the meridian. However, in the late 1840s Airy 
decided that a new and much larger meridian instrument 
was needed. So that there should be no break in the 
sequence of transit observations, the new transit circle was 
erected in the old Circle Room, east of Bradley's Transit 
Room, while regular transit observations - particularly 
necessary for time determination, for example - continued 
on Pond's transit instrument. 

On the first observing day in the new half-century, 4 
January 1851 (the methodical Airy would have liked this to 
have been 1 January, but English weather frustrated him), 
the new transit circle was brought into use-and this had the 
effect of moving the Greenwich Meridian some 19 feet to the 
eastward, a difference of less than '/so second in the time of 
transit, a quantity which at that date was too small to be 
measurable. When after 1884 Greenwich was chosen as Ihe 
world's prime meridian, one country which did not have to 
change its maps was Britain - or so one would assume. 
However, it seems that, although Airy did inform the 
Ordnance Survey of the proposed change in instruments in 
1850, they failed to change the records. 

When all this was taken into account, the discrepancy 
between the old and the new triangulations was reduced to 
6 centimetres in latitude and i .95 metres in longitude. 22 



A clock more accurate than the Earth 

Because this story is primarily that of the distribution and 
the uses made of time, we have so far made but passing 
reference to observatory clocks. Until recently the funda- 
mental timekeeper was the rotating Earth, and time was 
found at frequent intervals by astronomical observations; 
clocks were only used to 'keep' time in the comparatively 
short intervals between observations. In this chapter, how- 
ever, it is the developments in the clocks themselves - and 
the consequences of those developments - that form the 
main part of the story because, within the last forty years, 
man-made clocks have been developed which are better 
timekeepers than the Earth itself. 

In the first two hundred years of the Royal Observatory's 
existence there was some increase in the degree of precision 
of the pendulum clock due to the invention of the dead-beat 
escapement and temperature-compensated pendulum by 
Graham and others early in the eighteenth century, but 
these were not really fundamental developments. In 1676 
Flamsteed's year-clocks could be relied upon to about 7 
seconds per day; in 1870 Airy's barometrically compen- 
sated regulator clock (Dent No. 1906) was accurate to about 
0.1 seconds per day (in accuracy, somewhat ahead of its 
time). These and other developments in timekeeping are 
dealt with in rather more detail in Appendix III. 

In the last decade of the nineteenth century the leading 
astronomical observatories (though not Greenwich) began 
acquiring clocks significantly more accurate than their 
predecessors, designed by Siegmund Riefler of Munich 
(1847-1912). But it was not until the 1920s that the first real 
breakthrough occurred - the Shortt free-pendulum clock, 
one of the most important developments in timekeeping 


since the invention of the pendulum clock itself two 
hundred years before. The free-pendulum idea had been 
pioneered by R. J. Rudd as long ago as 1899: a practical 
system was perfected in 1921-4 by William Hamilton 
Shortt, a railway engineer, working in conjunction with 
F. Hope-Jones and the Synchronome Co. Ltd. In ordinary 
pendulum clocks the foee swinging of the pendulum, on 
which timekeeping accuracy depends, is interfered with by 
the need to sustain the pendulum's motion and to count the 
swings to tell the time. In a free-pendulum clock, these two 
functions are carried out by a subsidiary 'slave clock', 
allowing the master pendulum to swing quite freely except 
for a traction of a second each half-minute when it receives 
an impulse from the slave. Before this, the best clocks had 
an accuracy of about 1 second in ten days: the Shortts were 
accurate to 10 seconds in a year. Greenwich acquired its first 
Shortt Iree-pendulum clock in 1924 when Shortt No. 3 took 
over as Sidereal Standard. Other Shortts followed. Within a 
very few years, the free-pendulum clocks ousted for time- 
service purposes all the older clocks, some of which (by 
George Graham) had been in active astronomical use for 
nearly 200 years, and none of which (except a recently 
acquired copy of a Riefler) was less than 55 years olti. 

One of the consequences of the increase in accuracy of the 
primary timekeepers was a change of concept in the oper- 
ation of the Greenwich time service. When Airy instituted 
the service in 1852 he based the timekeeping on two 
Standard clocks, the Sidereal Standard and the Mean Solar 
Standard. Radio time signals, however, made it possible to 
compare clocks in other observatories the world over with 
great precision several times daily. Furthermore, Green- 
wich itself had many more very accurate clocks. This led in 
1938 to the abandonment of Airy's standard-clock concept 
in favour of taking the mean of several clocks, some keeping 
sidereal, some solar time, initially five at Greenwich 
and one at the National Physical Laboratory, Teddington, 
England (NPL), to which was added a year later one from 
Edinburgh. All were Shortt free-pendulum clocks. 

Quartz crystal clocks 

We must now consider for a moment concepts of time, in 




particular the difference between an instant of time (the 
'date', or the 'epoch'), and an hitcnml of time. Someone 
catching a train or aircraft is interested primarily in the 
instant; a boxing referee for example, in the interval. There 
is a third concept and that is the frequency of a periodic 
phenomenon - the number of cycles of this phenomenon 
per unit of time: the name of the unit of frequency today is 
the hertz (Hz), identical with the older unit of a cycle per 
second (cps). 

It was the search by telecommunications engineers for a 
reliable standard of frequency of electro-magnetic waves 
which gave rise to the development of the quartz crystal 
clock, destined to prove an even more significant innova- 
tion than the free-pendulum clock had been some ten years 
earlier. The quartz crystal was developed with the advent of 
radio broadcasting in the early 1920s, giving for the first 
time a highly stable radio-frequency source. The first quartz 
clock proper was described by Horton and Marrison of the 
USA in 1928. The first quartz clock at Greenwich was in- 
stalled in 1939, and was of a type developed by Dye and 
Essen at the NPL, with an accuracy of about 2 milliseconds 
per day. (A millisecond (ms) is one thousandth of a second; 
a microsecond (/xs) one millionth of a second.) There were 
plans to install further quartz clocks at Greenwich but these 
were postponed by the outbreak of war and the transfer of 
the time service to the magnetic observatory at Abinger in 
Surrey, which was thought to be safer from enemy action 
than Greenwich. A reserve time-service station was set up 
at the Royal Observatory in Edinburgh in 1941. Although no 
quartz clocks were available at Abinger initially, informa- 
tion from two at the NPL was transmitted daily and in- 
corporated, with the free-pendulum clocks, into the 'mean 

The needs of war required the British time service to 
achieve a tenfold increase in the accuracy of radio time 
signals, in connection with the development of radar, and 
particularly for precision air navigation systems. Arrange- 
ments were therefore made with the Post Office Radio 
Branch in 1942 for information from their own quartz clocks 
to be transmitted daily to Abinger. Their performance was 
so good that in 1943 the Shortt clocks ceased to be part of the 



'mean clock'. The quartz clocks, their errors determined by 
astronomical observations at Abinger and Edinburgh, be- 
came the primary standards on which the time service was 
based, while the observatory clocks became secondary 
standards used for the control of time signals. Then, in 1944, 
the control of the international time signal from Rugby was 
taken over by new quartz clocks at Abinger, as was, in 1949, 
the control of the BBC's six-pip signal. The time-service 
station at Edinburgh closed down in January 1946 and, 
shortly after, six quartz clocks were installed at Greenwich, 
though the headquarters of the time service remained at 
Abinger which had twelve quartz clocks. Accuracies had 
meanwhile improved to about 0.1 milliseconds a day. 
Meanwhile the astronomers had moved away from the 
smoke and street lights of Greenwich to the clearer air of 
1 Ierstmonceux in Sussex, where the observatory became 
known as the Royal Greenwich Observatory (RGO). The 
time service moved from Abinger to Herstmonceux in 

The non-uniform Earth 

All these improvements in precision drew attention to 
another problem, ably summed up by Sir Harold Spencer 
Jones, tenth Astronomer Royal, in 1950: 

The rotation of the Earth provides us with our fundamental unit of 
time - the day. Hie first requirement of a fundamental unit is that it 
should be constant and reproducible; the unit should mean the 
same thing to all men and at all limes. In taking the d^\ , or, more 
precisely, the mean solar day, as the fundamental unit, from which 
we derive the hour, the minute, and the second as subsidiary units, 
it has been implicitly assumed that its length is invariable or, in 
other words, that the Earth is a perfect timekeeper. ' 

That the Earth was not a perfect timekeeper had been postu- 
lated by Immanuel Kant as long ago as 1754, but to get the 
full story we must go back another sixty years. In 1695 
Edmond Halley, through an analysis of ancient eclipses, 
had come to the conclusion that the Moon's motion round 
the Earth was accelerating, and this was later confirmed by 
direct measurement. In 1787 Laplace showed that this could 
be explained by slow changes in the shape of the Earth's 
orbit, but in 1853 Adams pointed out that these orbital 



changes would account for only half the observed value of 
the Moon's acceleration. After much debate, it was event- 
ually shown that that part of the Moon's acceleration not 
accounted for by Laplace's gravitational theory could be 
explained by assuming a progressive slowing down in the 
rate of the Earth's rotation, due largely to the friction caused 
by the tides. 

We now know that there are three different types of 
changes in the Earth's rotation rate, the first two of which 
had been known for some time from studies of the motions 
of the Moon and planets; the last, though detected by free- 
pendulum clocks, was not evaluated until the advent of the 
quartz clock: 


1 second 
pet year 

Reference standard 

(based on the average 
length of the day in the 
nineteenth century) 


1 second 
per year 

1955 1980 

1955 1960 • 1965 1970 1975 1980 






49. The Earth .is ,i clock, showing changes in the Earth's rotation rate over the last JooyeaiS. 
The graph shows the three different aspects of this referred to on this page and the ne\t: 
( l ) long-term trend (or secular changes); (2) irregular (and unpredictable) fluctuations; (3) 
annual se.1son.1l fluctuations. 

49 (a) secular changes: the progressive slowing down already 

referred to, caused by tidal friction in shallow seas, amount- 
ing to an increase in the length of the day of about 1.5 
milliseconds per century; 

(/>) irregular (and unpredictable) fluctuations, probably 
caused by differing rates of rotation between the molten 



core and the solid mantle of the Earth, which may result in 
the length of the day increasing or decreasing by up to 4 
milliseconds in a decade; 

(c) seasonal variations, reflecting seasonal changes in the 
Earth's ocean and air masses. An example of this is the 
melting and freezing of the polar ice caps, and the move- 
ments of air masses due to the large area of high pressure 
over Siberia during winter being replaced by low pressure 
in summer. The Earth runs slow in the northern spring and 
early summer, fast in the autumn, the difference in the 
length of the day between the two extremes being in the 
order of 1 .2 milliseconds. 

There is another phenomenon which, though it does not 
affect the Earth's speed of rotation, has to be taken into 

0;1 02 03 04 05 seconds of latitude 


Towards 180° 

90° W 



55 Towards 

90° E 


50. Polar wander. 1971-7. 

Towards 0° (Greenwich) 

-I I l_ 


15 metres 




account when great timekeeping precision is needed. This 
is 'polar variation', or the wobble of the Earth on its axis 
(rather like the effect of a 'sloppy' bearing in a machine). 
This causes the Earth's poles to wander in an approximately 
annual circular path with a radius of about 8 metres. The 
effect of polar variation is to change the geographical 
latitude and longitude of every place on Earth (when as- 
certained by astronomical means) by a minute amount, the 
consequence of which is that there are minute variations in 
the time scale at each place due to changes of longitude. 

As Spencer Jones said, the first requirement for a funda- 
mental unit is that it should be constant and reproducible. 
By 1950 the second of time based upon the rotating Earth 
had been proved to have variations in length which, al- 
though insignificant hitherto, could no longer be neglected. 
What was to be done? 

Ephemeris time 

The first solution was to abandon the solar day as the 
fundamental unit and to substitute the year, the length of 
which, though not constant, can be predicted, decreasing 
by about 0.5 seconds per century. This led to the intro- 
duction internationally in 1952 of a new kind of time scale 
for certain purposes - Ephemeris Time (ET), which is used 
(as the name implies) in compiling the various national 
ephemerides and almanacs. But first we must say some- 
thing about Universal Time. As we saw in the last chapter, 
Greenwich Mean Time came to be known by astronomers 
and other scientific users as Universal Time (UT), as a result 
of the decisions of the Washington conference in 1884 and 
the specific recommendations of the 1AU in 1928: so in this 
chapter, we shall speak of UT rather than GMT when re- 
ferring to the mean solar time on the Greenwich meridian. 
Now, UT is based on the spin of the Earth on its axis and is 
the time-scale needed for celestial navigation. But as we 
have seen, the rate of spin is variable, so in 1956 it became 
necessary to define UT more closely for specialist use in the 
time services: 

UTo Mean solar time of the prime meridian obtained 
from direct astronomical observation. 

UTi UTo corrected for the observed effects of polar 


motion (maximum about 0.035 seconds). UTi is the time- 
scale used in celestial navigation. 

UT2 UTo corrected for observed polar motion and for 
extrapolated variations in the Earth's rotation rate (also up 
to about 0.035 seconds). UT2 is a 'smoothed' time-scale 
giving as uniform a time as possible, on which time signals 
before 1972 were based. 3 

The full story of ET and of its relationship to UT is too 
complex to tell here. Suffice it to say that ET was made to 
conform closely to UT by making the Ephemeris day the 
average length of the mean solar day during the nineteenth 
century. Finally, in 1956, the mean solar day was aban- 
doned internationally as the fundamental unit of time in 
favour of the Ephemeris Second, defined as 'the fraction 
1/31 556 925-9747 of the tropical year for 1900 January cA 12^ 
ephemeris time'. 4 

This change, however, did not solve all problems. Be- 
cause it was invariable, a second defined in terms of 
Ephemeris Time suited many theoretical needs and was 
immediately adopted for the various ephemerides. But it 
was not suitable for everyday use for two main reasons: 
first, it was not readily accessible and could only be deter- 
mined to the required accuracy after a long delay, several 
years' worth of observations having to be analysed before 
any kind of result could be obtained; secondly, those who 
were interested in the precise instant of time as opposed to 
the time interval - and these included the public at large - 
required that time signals should conform fairly closely to 
the rotation of the Earth, to the alternation of night and day. 
Though the differences between ET and UT were very small 
in any one year, the accumulated differences could become 
significant because of the systematic slowing down of the 
Earth's rotation. By 1952 when ET was first brought into 
use, there was an accumulated difference of almost 30 
seconds between ET, based on the nineteenth-century 
rotation rate, and UT, based on 1952. 

The use of Ephemeris Time in time signals proved at best 
to be a compromise because the physicist and telecommuni- 
cations engineer required that the length of the second on 
their time signal should be uniform - that it should 'mean 
the same thing to all men and at all times' - while the 



ordinary user (and the navigators and surveyors) required 
that when a time signal said it was noon, then it should be 
noon according to the heavens. Before 1944 time signals 
controlled by Greenwich had been kept as closely as 
possible in line with the Earth's rotation, resulting in a 
second (as obtained from time signals) that could vary in 
length from day to day, albeit very minutely. Then, from 
1944, an effort was made to transmit time signals at as 
uniform a rate as possible, based on the mean of the best 
quartz clocks then available, with 'jump' corrections in- 
troduced when necessary (on Wednesdays) to maintain 
agreement with universal (astronomical) time. In the USA 
at that date, however, no such attempt at a compromise 
between time and frequency was made: time signals from 
radio station NSS (Annapolis), controlled by the US Naval 
Observatory, were kept strictly in line with the Earth's 
rotation while standard frequency transmissions from radio 
station WWV controlled by the US National Bureau of Stan- 
dards were kept at as uniform a rate as possible. 

The atomic clock 

The answer to the first of the disadvantages of Ephemeris 
Time, inaccessibility, proved to be the atomic clock. The first 
operational complete atomic clock system was developed at 
the US National Bureau of Standards, Washington, DC, by 
Harold Lyons and his associates in 1948-9, using an 
absorption line of ammonia to stabilize a quartz crystal 
oscillator. First brought into operation on 12 August 1948, it 
was developed specifically for use as a frequency standard. 
However, attention soon turned away from ammonia to- 
wards the element caesium. Most of the early development 
of the caesium standard, with which the names of ]. E. 
Sherwood, J. R. Zacharias, and N. Ramsay are particularly 
associated, took place in the USA. But it was at the National 
Physical Laboratory, England, that a caesium beam stan- 
dard was first used on a regular basis. Designed by L. Essen 
and J. Parry, this was brought into use for the calibration of 
quartz clocks and frequency standards in June 1955, the 
very year that the decision was taken that the ephemeris 
second, with all its limitations, should be adopted as the 
fundamental unit of time. During the next few years further 



laboratory- type caesium standards were brought into use in 
Boulder (Colorado), in Ottawa, and in Neuchatel. s 

Even these earliest examples of atomic clocks were a 
hundred times more stable in the long term than the best 
quartz crystal standard and did not suffer from the drifts in 
rate associated with the 'ageing' of quartz crystals. De- 
velopments from them have now provided a time-scale of 
great uniformity, of very high precision (at least ten times 
more accurate than the prototype), and of almost immediate 
accessibility. But it was some years before all this could 
be realized. Only the latest super-tube atomic caesium 
standards achieve a short-term stability comparable with 
that of a free-running quartz clock. 

Seconds per day, 
gain or loss 

100 000 000" 


1 microsecond Tooooojj 


100 000 




1 millisecond 

1 second 





Improved caesium 

NPL caesium atomic 

Quartz crystal 
I Shorn free pendulum 
Riefler/lmproved escapement 
I Airy/' Barometric compensation 

Harrison & Graham/Tomporaturo compensation 
Graham /Dead-beat escapement 
Tompion/lmproved movement 

1700 1800 1900 7000 AD 

51 . The increasing accuracy of precision clocks (order of magnitude only). 

All clocks have to be adjusted so that they go at the correct 
rate- that they 'keep time' -and also have to be set to time. 
The new atomic clocks were no exception and the first task 
was to calibrate them against accepted standards; in other 
words, the atomic time-scale had to be related to the astro- 
nomical time-scale. Between 1955 and 1958 atomic clocks in 
the UK and USA were calibrated against astronomical time- 
scales at Herstmonceux and Washington. The first atomic 
time-scale was known as GA (Greenwich atomic), using 


Time to gain or 
lose 1 second 

30 000 years 
-| 3000 years 

300 years 

30 years 

3 years 
1 year 
3 months 

10 days 

1 day 
3 hours 


initially the NPL caesium standard with a rate set to agree as 
closely as possible with F.phemeris Time. Then in 1959 the 
US Naval Observatory's time-scale Ai was brought into 
general world-wide use. Its point of origin was arbitrarily 
set so that Atomic Time and UT2 were the same at midnight 
on 1 January 1958. At the same time the atomic second was 
defined in terms of the resonance of the caesium atom. In 
1964 the atomic second was recognized internationally as a 
means of realizing the ephemeris second. Then, at the 13th 
General Conference of Weights and Measures in Paris in 
1967, the astronomical definition of a second was aban- 
doned and the atomic second adopted as the fundamental 
unit of time in the International System of Scientific Units 
(SI system), with the following resolution: 

That the unit of time in the International System of Units shall be 
the second, defined as follows:- The second is the duration of 9 192 
631 770 periods of the radiation corresponding to the transition 
between the two hyperfine levels of the ground state ot the 
caesium- 133 atom. 6 

As other caesium clocks came into use all over the world - 
clocks which could be compared with each other by radio 
and other means to an accuracy of a microsecond or better - 
it became possible to form an international 'mean clock' of 
great precision, the larger number of independent contri- 
butions achieving very great uniformity. These clocks 
agreed with each other to within a few microseconds in a 
year, whereas they differed from the time-scale of the rotat- 
ing Earth (Universal Time) by up to a second in a year. 

The Bureau International de I'Heure which has co- 
ordinated international timekeeping since 1919, had formed 
its own atomic time-scale (A3) soon after that of the USA, 
based on the results from three independent standards in 
England, Switzerland, and the USA, with its origin on 1 
January 1958. The BIH atomic time-scale was formally 
adopted internationally in 1971, the name being changed to 
International Atomic Time (TAI). It is worth noting that, on 
i January 1979, twenty-one years after the two scales had 
arbitrarily been set together, UTi (based on the Earth's 
actual rotation rate 1958-79) had lost some seventeen 
seconds on TAI (based on the Earth's rotation rate in the 
nineteenth century). 


Co-ordination of time signals 

To return to time signals, in 1958 the UK time service 
adopted a system later called Co-ordinated Universal Time 
(UTC), the aim of which was to keep the signals within 
approximately a tenth of a second of UT2. This was 
achieved by slightly changing the rate of the atomic clocks 
generating the time signals, applying an 'offset' which 
caused them to approximate to the current rate of UT2. (In 
the 1960s this was a losing rate in respect of Atomic Time.) 
The offset value remained unchanged through each calen- 
dar year but, to cater for unforeseen changes in the Earth's 
rotation rate, step corrections were applied each month to 
keep UTC within 0.1 seconds of UT2. In 1961 full co- 
ordination between UK and US time services was insti- 
tuted. The time signals were synchronized, and the same 
annual offsets and monthly step corrections applied. In 
1963 this UK/US system was extended world-wide, being 
operated under the control of BIH in Paris. It was then 
that the name Co-ordinated Universal Time was adopted. 7 
I lowever, the increase in number and sophistication of 
satellite and other electronic communication and navigation 
systems brought to light major practical difficulties. These 
systems depend upon the very precise synchronization of 
both the radio signals themselves and of frequencies. The 
resetting in time in line with the step corrections proved 
difficult and inconvenient; the resetting in frequency once a 
year even more so. The fact that the second as broadcast in 
time signals did not conform to the legal second was an 
aesthetic rather than a real difficulty. 

The leap second 

After much discussion at all levels, national and inter- 
national, fundamental changes were made in the system of 
time signals world-wide. As from 1 January 1972 time 
signals radiated atomic seconds precisely, the new UTC 
being set to be exactly 10 seconds slow on TAI as from that 
date. This is the system in operation at the time of writing. 

It was originally agreed that time signals should not be 
allowed to depart more than 0.7 seconds (a tolerance later 
increased to 0.9 seconds) from UTi, the time-scale needed 
by navigators and astronomers. To achieve this, step correc- 





tions, either positive or negative, of exactly one second, 
known as 'leap seconds', are made when needed on the last 
day of a calendar month, preferably 31 December and/or 30 
June. Just as the fact that there is not a whole number of 
days in a year gave rise to the leap year, so the fact that there 
is not a whole number of atomic seconds in a solar day gave 
rise to the leap second. 

Thus international time and frequency radio signals, 
such as those from station MSF in the UK or WWV in the 
US, radiate exact atomic seconds without interruption or 
variation throughout the year. All that happens when a leap 
second is added or subtracted is that the numbering of the 
seconds markers is changed. Thus, if a positive leap second 
is needed on 3 1 December because UTC is seen to be drifting 
too far from UTi, the last 'minute' of the year will have 
sixty-one seconds instead of the conventional sixty. For a 
negative leap second, the last 'minute' would have only 
fifty-nine seconds. To cater for those who must know UTi 
more precisely (navigators and astronomers for example), 
the principal time and frequency signals contain a form of 
code which indicates the number of tenths of a second that 
the time signal (UTC) on that particular day is fast or slow on 
astronomical time (UTi). 

The world's time signals are now co-ordinated by the 
BII I in Paris, based on a world 'mean clock' comprising 
some eighty atomic clocks from twenty-four participating 
countries. (Only those countries within the coverage of the 
radio navigational aid l.oran-C are at this time able to par- 
ticipate, but future satellite navigation systems will allow 
many more clocks to be compared.) It is the F51H who decide 
when leap seconds should be inserted. In 1972 TAI was set 
to be precisely 10 seconds behind UTC since then, eight 
positive leap seconds have had to be inserted. At 1 January 
1979, therefore, UTC lagged exactly 18 seconds behind TAI 
which always differs from UTC by an exact number of 

With the introduction into time signals in 1972 of the new 
UTC tied to atomic time (TAI), in place of the old UTC tied to 
mean solar time (UT2, which most non-scientists continued 
to call GMT), theargumentsabout the nomenclature of time- 
scales started again. Of course the new time-scale continued 


to be based on the Greenwich meridian, but it could no 
longer be defined as mean solar time on the Greenwich 
meridian (which is GMT) even though it might never depart 
more than nine-tenths of a second from it. Indeed, even the 




I I 


o 10 





1972 Jan 1 , 

when UTC was set to be 

precisely 10 seconds 

slow on TAI, and when 

the leap-second system 





52. Time signals and the leap second. 

Greenwich meridian itself is not quite what it used to be - 
defined by 'the centre of the transit instrument at the 
Observatory at Greenwich'. Although that instrument still 


survives in working order, it is no longer in use and now the 
meridian of origin of the world's longitude and time is not 
strictly defined in material form but from a statistical solu- 
tion resulting from observations of all time-determination 
stations which the BIH takes into account when co-ordinat- 
ing the world's time signals. Nevertheless, the brass line in 
the old observatory's courtyard today differs no more than a 
few metres from that imaginary line which is now the prime 
meridian of the world. 

Though no longer in use in astronomy proper, the term 
GMT continued to be used by navigators, for many civil 
uses, and as a description of statutory legal time in many 
countries. However, there was opposition to the continu- 
ance of this state of affairs, particularly from France. In 1975 
the 15th General Conference of Weights and Measures 
(CGPM) recommended to its adhering countries that, as the 
new UTC was used in time signals, it should in future be 
used as the basis of legal time: 8 it would thus replace GMT 
which, because of the change in derivation of UTC in 1972, 
was now ambiguous. 9 France and Spain have already taken 
legislative action; Holland, Switzerland, and the Federal 
Republic of Germany are, at the time of writing, in the 
process of doing so. France issued a decree on 9 August 
1978, repealing the law of 1911 (which said that legal time in 
France was Paris mean time retarded by 9 minutes 21 
seconds) and ordaining that legal time in all French terri- 
tories should in future be obtained by adding or subtracting 
an exact number of hours to UTC, which can be increased or 
reduced during part of the year (thus allowing for Summer 
Time), and that the term GMT was not to be used in any 
future regulations. "' 

Because 1 atomic second had to be added to the year 1978, 
it might be thought that 1978 was a second longer than the 
previous year. That is not so. So far as is known, the length 
of the year is decreasing only by some half a second per 
century. It is the day - the Universal Day (and the hour, and 
the minute, and the second) - that is getting longer. Thus, 
365 1978-type days were longer by about one second than 
365 nineteenth-century-type days on which the time signals 
are based. One leap second had to be added to 1978 to make 
certain that, for the early part of 1979 at least, the noon 'pip' 


of the time signal occurred within 0.9 seconds of noon 
according to the stars. 

It is not possible to predict how Earth's rotation rate will 
change over the next few decades. At present, it is slowing 
down considerably faster than the average over the last 
three hundred years. However, it could be that the trend 
will be reversed and that, should the present system of time 
signals persist, no leap second, or even negative leap 
seconds, will have to be inserted by, say, the 1990s. Never- 
theless, it is inevitable that, some time in the future - per- 
haps in tens of years, perhaps in hundreds, perhaps even in 
thousands - there will be a need for two and then three 
positive leap seconds each year, should we continue to base 
our time-scale on the average length of the day in the 
nineteenth century. As for the more distant future, one of 
the effects of the decelerating Farth will be that the need for 
leap years (but not for leap seconds) will disappear: in a few 
million years' time, there will only be 365 days in the year, 
not 363 '/a as now. 


In this book we have told how, when the Royal Observatory 
was founded three hundred years ago, Greenwich time 
concerned only one astronomer and his assistant; how in 
the 1760s the publication of The Nautical Almanac and the 
development of the marine chronometer meant that the 
Greenwich meridian and Greenwich time began to be used 
by mariners of all nations; how, ashore, at the turn of 53 

the nineteenth century,, the increasing use of clocks and 
watches led to the abandonment of apparent, or 'sundial', 
time and the adoption of mean, or 'clock', time as the local 
time kept in each community; how, not many years later, 
the development of railways led to the abandonment of 
local time and the adoption in each country of a national, or 
'railway', time; how, by 1884, the development of world- 54 

wide communications was such that there was a need for 
international, or 'universal', time, and how it was Green- 
wich time that was chosen for this both ashore and afloat; 
how in the 1940s and 1950s the development of quartz and 
atomic clocks (the latter more accurate than Flamsteed's 
pendulum clocks by a factor of some 8 million) resulted in 

54- The Greenwich time-ball today 


the adoption of atomic time and the abandonment of the 
Earth as the fundamental timekeeper, the Earth's rotation 
rate having proved, after all, not to be uniform. 

Britain's time signals no longer emanate from Greenwich 
itself. The world's time is now co-ordinated from Paris (and 
is called Universal Time). Universal Time today truly is 
universal, being based on the mean of clocks from twenty- 
four different nations. Nevertheless the world's prime 
meridian for longitude and time still passes through the old 
55,56 observatory at Greenwich. Although the world's time 

signals (and in some countries legal time also) are no longer 
based on GMT as strictly defined, nevertheless the time- 
scale on which they are based - UTC - is bound to be within 
a second of the old GMT. 


i Finding the longitude 


The difference in longitude between any two places on the surface 
of the Earth is precisely equal to the difference between the local 
times of the two places. It is this unalterable relationship which has 
made it necessary for a book such as this, primarily about time, to 
have a sub- title 'the discovery of the longitude'. Furthermore, until 
recent times, many of the most important developments in time- 
keeping have been directly stimulated by the need to find longi- 
tude at sea. The purpose of this appendix is to explain in simple 
terms why this should have been so: it should be read in conjunc- 
tion with Chapter i which gives the historical facts. 

It is explained in the first chapter that geographical positions 
have been described in terms of latitude and longitude at least since 
the time of Ptolemy in the second century AD. finding latitude has 
been possible since ancient times, by making some sort of 
measurement of the Sun's altitude at noon, at first by measuring 
the lengths of shadows of a gnomon of known height, then by 
more sophisticated instruments developed for the purpose, first 
ashore and then, from the fifteenth century, at sea too. Latitude 
could be found also by measuring the height of the Pole Star above 
the horizon, although an allowance had to be made for the fact that 
the star is not situated exactly at the celestial north pole (which in 
northern latitudes is always elevated above the horizon at precisely 
the amount of the observer's latitude). 

A practical method of finding longitude, however, did not come 
within reach until the last few hundred years. Nevertheless, the 
theoretical solution is comparatively simple and was known to 
Hipparchos, for example. When the Sun is on your meridian and it 
is 12 noon to you at C, it is 6 a.m. (and at the equinox it would be 
sunrise) for person A situated on a meridian 90" 'msl of yours, while 
it would be 9 a.m. for person B 45° to the west. Similarly it would be 
3 p.m. for person C 45° ens/ and 6 p.m. for person D 90" east. An 
hour later at 1 p.m. forG, it would be 7 a.m. for A, 10 a.m. for 8,4 
p.m. for C and 7 p.m. for D: though the times have changed, the 
differences of times remain constant and are a measure of the 



differences of longitude between the various places. Hipparchos 
probably thought of difference of longitude in terms of time, of so 
many hours east or west: today, we usually think of it in terms of 
arc, of so many degrees, minutes, and seconds east or west; but in 



12 midnight 







57. I ongitudo and time (the Earth shown .>- at the equinox). 

this context, time and arc are completely interchangeable, 24 hours 
answering to 360°, 12 hours to 180 3 , 1 hour to 15°, and 4 minutes to 
i c . If, therefore, you know that, when it is 9 a.m. to you. it is at that 
precise moment 7 a.m. for person £, then the difference of longi- 
tude must be 2 hours, which is equivalent to 30 ; because it is earlier 
in the morning for £ than for G (and as the Sun appears to move 
from east to west), then E must be 30 west of G. 

But how does person G know what time it is at, say, £, over a 
thousand miles to the west? One way of finding out would be to 
transport some form of timekeeper between the two places, but 
this did not become technically possible until the eighteenth cen- 
tury. Another method would be to record the time of some 
phenomenon which could be seen at both places simultaneously; 
then, at some later date, the time recorded at the two places could 
be compared and the difference of longitude found. A refinement 
of that method would be to predict the time at which the 
phenomenon would occur at some chosen place (say Greenwich) 
and then the longitude difference could be found immediately 
from the local time of the observed phenomenon without having to 
wait to compare actual measured times. 


Longitude by lunar eclipse 

As we saw on p. 2 above, Hipparchos is said to have proposed the 
use of a lunar eclipse as the phenomenon to be observed - when 
Sun, Earth, and Moon are directly in line and the Earth's shadow 
moves across the Moon's surface, reaching any particular point at 
precisely the same instant for an observer wherever he may be. 
Ptolerrvj recommended this method of finding longitude ashore 
(though he gives only one example of its having been used) but lie 
does not tell us how the local time of the phenomenon should be 
found. For the shadow to be seen on Earth at all, the Sun must be 
below the horizon, so a sundial could not have been used directly. 
and only the most rudimentary timekeepers would have been 
available - perhaps a water-clock, or an hourglass. 

This is still something of a mystery. One could use the position of 
the stars, either by measuring the position angle of the Great or 
Little Bear in relation to the Pole Star (as was to be done with the 
nocturnal, an instrument invented over a thousand years later), or 
by measuring the altitude and finding the time by some form of 
planispheric astrolabe. Or was the Moon's shadow used in some 
way, remembering that it would be Full Moon so that, in the 
middle of the eclipse, the Moon would be exactly opposite the Sun? 
A Spanish method of 1582 for use in the Indies recommended the 
setting up ashore of a vertical gnomon precisely one- third of a yard 
long: the length of the Suit's shadow when at its shortest the day 
before (or day after) the eclipse gave the latitude; the direction and 
length of the Moon's shadow at the beginning or end of the eclipse 
could (after considerable calculation, generally carried out back in 
Spain) give the longitude. There are few reports of this elegant 
method having been used but, from an observation of an eclipse oi 
1584, the position of the Casa Real in Mexico City was calculated as 
being only 21 km too far south and 23 km too far west of the true 
position, a remarkable result even taking into account a large 
element of luck. ' 

Longitude by lunar distance 

There were no fundamental developments in the longitude story 
until 1514 when, as we saw on p. S, Johann Werner described the 
lunar-distance method, which was the first to give some hope ol 
finding longitude tit set. During the course of a month, the Moon 
appears to move one complete revolution from west to east against 
the background of stars. This comparatively fast movement - 
approximately '/j°, or her own diameter, in an hour - prompted 
Werner to suggest using the Moon as a gigantic celestial clock, with 
the Moon as the hand and the stars in the zodiac as the figures on 
the dial . As seen from an observer on Earth on any particular night, 

appendices 195 

the stars in the zodiac move across the sky from east to west at a 
fraction more than 15° per hour, while in the same period the Moon 
moves about 14' :'. Thus the Moon appears to lag behind the stars 
at the rate of about half a degree per hour: a zodiacal star which is, 
say, 40° ahead of (that is, to the west of) the Moon at 22.00 will be 
41° ahead at midnight (in fact, the star will have moved about 30 
while the Moon moves 29°). Werner assumed that the 'lunar dis- 58 

tance' between zodiacal star and the Moon at a particular instant 
would be the same whatever the observer's geographical position. 

Sun's declination 


18° S 

58. Lunar-distance changes, relative to the Sun (as illustrated here) or to 
zodiacal star.(Quantities for 1 Oct. 1772, fig. 17.) 

and he therefore proposed that the changing lunar distance could 
be used as a measure of the time on some reference meridian 
(which we will henceforth call Greenwich). This could then be used 
to find longitude, provided the Moon's position with reference to 
the stars could be predicted in advance, and the actual distances in 
the sky measured by the navigator to the required degree ol accu- 

Werner's assumption that the Moon's position in the sky is not 
affected by the observer's geographical position is fallacious. 
Making allowance for this - parallax - and for the different 
amounts of refraction affecting the two bodies if they are at dif- 
ferent altitudes, was to prove one of the most tedious pieces of 
arithmetic which the navigator eventually had to perform. 
Werner's basic method was, however, perfectly sound in theory, 
although there were three fundamental needs to make it work in 
practice: knowing the precise positions of the stars relative to each 
other; being able to predict several years ahead the Moon's position 
against that frame of reference (the navigator had to have these 
predictions before he sailed, and voyages could last several years); 
and having an instrument which could take the necessary observa- 
tions to the required degree of accuracy. It was 250 years before 
these needs were satisfied. 




Lunar distances became a practical possibility in the 1760s with 
the publication of the Nautical Almanac where the first two require- 
ments were met by the tables of distances from the Moon to the 
Sun and certain zodiacal stars, predicted in the almanac for every 
three hours throughout the year. The third requirement was met 
by the sextant. Briefly, a lunar-distance sight required three simul- 
taneous or near-simultaneous observations - the angular distance 
from the Moon to a star or to the Sun, the altitude of the Moon, and 
the altitude of the star or Sun, the times being taken by the best 
watch available. Though there were advantages in taking observa- 
tions when the horizon was fully visible (in daylight or at twilight), 
this was not essential because great precision was not absolutely 
necessary for the altitude observations; so long as some horizon 
could be seen, this would suffice. The observations would then be 
'reduced' as follows: 

Step A - finding the local timeofobsersation: from an altitude observa- 
tion of Sun or star, preferably that taken at the same time as the 
lunar distance, but, if there had been a bad horizon, it could have 
been from an altitude taken the previous afternoon or the next 
morning, carried forwards or backwards by the watch: the local 
time was found by relatively simple spherical trigonometry; 

4- ®Lw»r distance 


m*m ■ i ■^ r "- :r: - ■•'■' 

59. Lunar-distance observations, showing the three near-simultaneous 
observations required to find longitude by lunar distance: (1) the angular 
distance between the Moon and selected star (or Sun); (2)<iltitudeof Moon 
above horizon; (3) altitude of star or Sun above horizon 

SlepB - clearing the observed lunar distance from the effect* of refraction 
and parallax: tables were provided for doing this, but the calcula- 
tions were laborious; the end-product was the corrected lunar 
distance at the time of observation between the centre of the Moon 

appendices 197 

and the centre of the Sun or star, as seen from the centre of the 59 

Larth - which is the form in which lunar distances are tabulated in 
the almanac; 

Step C - finding the Greenwich time of observation: by entering the 
corrected lunar distance obtained in step B in the lunar distance 
table in the almanac, interpolating as necessary. For example, 
about 04.30 astronomical time ( 16.30 civil time) on 4 October 1772, 
William Wales, astronomer in Captain Cook's ship Resolution in the 
South Atlantic while on passage from Plymouth to the Cape of 
Good I lope, took four consecutive lunar-distance observations of 
the 'Limbs' (edges) of the Sun and Moon, the mean of which was 
102 26' 55". The corrected lunar distance (Step B) came to 102° 36' 
08". Then, interpolating in the lunar distance table in the Nautical 
Almanac between the value given for 6 h and that for 9 h (for the 17 

Sun on 4 October on p. 65 above), the Greenwich time of the 
observation proves to have been 6 h 23 m 39 s.* 

Step D - finding the longitude: by taking the difference between the 
local time found in Step A and the Greenwich time found in Step C. 
Thus, taking the same example: 

h m s 
Local time from Step A 04 26 56 

Greenwich time from Step C 06 23 39 

Longitude in time 01 56 43 

= Longitude in arc 29 1 1 ' West 

(To find whether cast or west, repeat the jingle 'Longitude east, 

Greenwich time least; longitude west, Greenwich time best.' The 

example is given to full numerical accuracy but an error of 10" in the 

observed distance corresponds to a minimum error of about 5' in 


Longitude by eclipses of Jupiter's satellites 

The invention of the telescope in the first few years of the seven- 
teenth century revealed the existence of four bright satellites 
revolving around Jupiter: like our own Moon, they were eclipsed 
periodically when they passed into the planet's shadow. Galileo 
pointed out that, if these could be predicted and the eclipses 
accurately timed, they could be used for finding longitude exactly 
as with the eclipses of our own Moon, but with the added advan- 

*Board of Longitude papers (RGO MSS.) vol. 53, f.46 v , where this and 
other lunar-distance observations taken by Wales in 1772 are preserved. 
They are written in manuscript in a book of printed forms designed by 


tage that they occurred much more frequently and the duration of 
observation was much shorter. This method was used extensively 
ashore but proved impracticable at sea because of the difficulty of 
observing. Tables giving predicted times are given in the almanacs. 

Longitude by chronometer 

As described in Chapter 3, this became a practical possibility in the 
second half of the eighteenth century. The basic principle for 
finding longitude was similar to that described for lunar distances 
above, except that one altitude observation only is needed and, in 
the reduction, only Step A was needed. I his considerably reduced 
the labour involved; furthermore, the method is potentially far 
more precise. 

11 Time-finding by astronomy 

The method 

For astronomical purposes, the basic requirement in lime-finding 
is the determination of the precise moment of mean noon, that is, the 
moment the Mean Sun is at its highest, or culminates, which occurs 
when it crosses, or transits, the observer's meridian. The interval 
between two successive transits of the Mean Sun across the same 
meridian is a Mean Solar day, which can then be divided into luntrs, 
minutes, and seconds by some form of timekeeper such as a sundial 
or clock. The Mean Sun is a fictitious body, created by astronomers 
when it was realized in Hellenistic times that the real Sun was not 
the best of timekeepers, sometimes running a little fast, sometimes 
a little slow. Of course, until the coming of accurate clocks in the 
last few hundred years, the concept of mean solar time was of 
interest only to astronomers: for the purposes of ordinary life, it 
was the real Sun - its rising, its culmination, its setting - which 
governed the time of day, and it was the sundial which told the 
time - apparent solar time - when needed. 

I Jere, however, we are concerned with astronomers and there- 
fore with mean rather than apparent time. The Mean Sun being a 
fictitious body which moves around the celestial equator at a con- 
stant speed (whereas the real Sun moves around the ecliptic at 
varying speeds in the course of the year), the moment of mean 
noon cannot be determined by direct observation but only 
indirectly through observations of some real body. lor obvious 
reasons, it was observations of the real Sun which were used lor 
this purpose in earlier times, the instant of mean noon being then 
obtained bv applying the equation of time, which is the difference 
between mean and apparent solar time at any moment. Depending 
as it does upon the Sun's declination north or south of the celestial 
equator, and upon the Sun-Earth distance on the day concerned, 
the size of the equation varies through the year. On 4 November 
each year, for example, the real Sun crosses the meridian some 16 
minutes before mean noon, while on 2 September the equation is 
zero and mean and apparent times are the same. 



Bul using the real Sun for precise time determination presents 
several disadvantages: the Sun is not easy to observe accurately 
and it may be cloudy at the critical moment. Astronomers, there- 
tore, seldom use the Sun for time determination, preferring instead 
to observe certain bright stars whose positions are very accurately 
known. Direct observations of any one of these will yield sidereal 
lime, which can be converted to mean solar time whenever needed. 
The position of a star which appears to the observer as a point 
source of light can be determined very precisely. As there are many 
stars, cloudy weather is less of a hazard and the overall precision of 
time determination is increased because a dozen or more observa- 
tions can be taken each night (and, for the brighter stars, in day- 
light as well), whereas it is generally not possible to take more than 
one solar observation for time determination each day. 

The instruments 

At the time Greenwich Observatory was founded the usual 
method of time-finding was by the Double Altitude, or Equal 
Altitude, Method, not to be confused with the later Gaussian 
method which is sometimes called by the latter name. In the 
Double Altitude Method the astronomer, using a movable quad- 
rant mounted on a vertical axis, took an altitude observation of the 
Sun when the altitude was increasing, an hour or so before noon, 
noting the time by the clock being checked. Then, at an equivalent 
interval after noon - the Sun having crossed the meridian and with 
its altitude now decreasing he noted the precise time the Sun 
reached the same altitude as in the forenoon observalion. With a 
small correction to allow for the change in the Sun's declination 
between observations, the time halfway between the two equal- 
allitude observations was apparent noon, from which mean noon 
could be found by applying the equation of time lor the day. This 
was the principal method used at Greenwich from 1676(0 1725, the 
various quadrants being set up in the Great Room. 

The invention of the transit instrument, which in its various 
forms was used for time determination at the Royal Observatory 
from 1721 to 1957, is ascribed to Ole Romer, the Danish astron- 
omer who first used such an instrument in Copenhagen in 1689. The 
first to be used in England was set up by Edmond Halley in 1721 at 
Greenwich, where it can still be seen. The simple transit instru- 
ment consists of a telescope fixed at right angles to a horizontal axis 
which is free to rotate upon pivotson two fixed piers. The telescope 
can be moved up and down but not from side to side and, because 
the axis is adjusted so as to be horizontal and precisely east and 
west, the line of sight of the centre of the field of view (the technical 
term is line of collimation, or optical axis) will always be in the plane 



of the observer's meridian. In principle, therefore, the instant any 
heavenly body crosses the meridian can be found from a single 
observation. In practice, the transit telescope has several horizontal 
and vertical 'wires' - actually they are usually spider's threads set 
at the telescope focus - which can be seen in the field of view. It 





60. A transit instrument, used in conjunction with a clock to determine the 
exact moment ,1 heavenly body crosses the meridian. The axis AP is 
horizontal and points east and west, so thai as the telescope CD swings on 
the axis it moves in Ihe meridian 

is essential that the optical axis of the transit telescope should 
always be directed towards the meridian. To ensure this, the transit 
observer has to take account of three possible errors, preferably 
before each set of observations - either to adjust the apparatus to 
remove them, or to know their magnitude so that they can be 
allowed for arithmetically: 

(<i) azimuth error, when the pivots on which the axis rotates are not 
exactly east and west: checked by observing <m azimuth mark set 
precisely on the meridian a mile or so distant (or by observing 
circumpolar stars) and corrected by moving one of the pivots 
backwards or forwards; 

(/') lavl error, when Ihe pivots are not horizontal: checked by a 
bubble level or nadir observations (looking vertically downwards 
into a bowl of mercurv) and corrected by raising or lowering one of 
the pivots; 

(c) collimation error, when the optical axis of the telescope is not at 
right angles to the rotation axis of the pivots: eliminated if the 
telescope can be reversed on its bearings during an observation. 



Simple transit instruments were used for time-finding at Green- 
wich from 1721 to 1850' but it must be remembered thai the transit 
instrument (and its development, the transit circle) is used exten- 
sively for the converse problem: given the time, find the right 

Airy's transit circle - which can measure declination as well as 
right ascensions and whose optical axis defines the world's prime 
meridian M as used for time-finding from 1851 101927. lime was 
recorded automatically by chronograph from 1854, and manv other 
refinements were introduced over the years, I lowever, the increas- 
ing precision in timekeeping brought about by the free-pendulum 
clocks gave rise to the need tor a corresponding increase in preci- 
sion in time determination. The telescope and axis of Airy's transit 
circle weigh nearly two thousand pounds and cannot be quicklv 
reversed on the pivots. From 1927, therefore, time determination 
(but not position measurement) at Greenwich (and during and 
after the Second World War, at Abinger, Fdinburgh, and Herst- 
monceux as well) was taken over by a succession of small reversible 
transit instruments (with telescopes about three feet long) on 
special stands which allowed the transit telescope to be reversed 
during an observation, thus eliminating collimation error. 

But, as we have seen, timekeeping precision increased fast in the 
1940s and 1950s, first with the quartz crystal and then with the 
atomic clock, demanding yet more precision in time deter- 
Light from star 

61. Photographic zenith tube general 
arrangement. The object glass I. is 
mounted on the rolarv, R; M is the 
mercury bath, P the photographic plate, 
and T the conical tube supporting the 
rotary, which turns in the ball race 15 
From ( kcaskmal Notes of the RAS, 2 1 Not 
19-59. 22 8 (by permission). 

63. The Greerm idl PZT Console. I here is no 
'observer' as Mich, the whole operation of the 
zenith tube being remotelv controlled from this 

62. The Greenwich photographic zenith tube. 

64. A Danjon impersonal prismatic astrolabe at 
Paris Obscrvatorv. 






mination. In 1957, me observatory having moved to Herst- 
monceux, time determination was taken over a Photographic 
Zenith lube (PZT) designed by D. S. Perfect.- The PZT is a 
development of the Reflex Zenith Tube devised by Airy about 1850. 
Airy's design was subsequently greatly improved and adapted for 
use in time determination at the US Naval Observatory. As the 
name implies, the telescope of the PZT is fixed so as to point 
vertically towards the zenith (through which the meridian passes). 
Its use is limited to measuring the transits of stars passing within 
15' of the zenith but this drawback is offset by the superior accuracv 
of the measurements. Kays from a star are reflected upwards from 
the mercury surface which defines the vertical, and the images 
from four successive exposures are automatically recorded on a 
moving photographic plate set centrally under the lens. The clock 
times of each exposure are accurately (and automatically) recorded 
by chronograph and the times at which the stars transit the mer- 
idian can be obtained from the positions of the images on the plate 
and the clock times at which they are observed. To obtain GMT 
(strictly, UTo), the local time obtained has to be corrected for the 
amount Herstmonceux is to the east of Greenwich - i m 2i\o785. 

Another instrument that has been used regularly at the Royal 
Greenwich Observatory and is used extensively for time deter- 
mination in many other observatories is the impersonal prismatic 
astrolabe invented by Andre Danjon and used for the first time in 
1951 at Besancon.' It is an instrument which can determine very 
precisely when a star reaches a fixed zenith distance of 30 and it 
has a rotary mounting so that it can be pointed in any direction. 
As with the PZT, the vertical is defined by reflection from a bowl of 
mercury. The transit instrument and PZT make use of the apparent 
motion of the stars across the meridian, whereas the prismatic 
astrolabe makes use of the motion of several stars across the almu- 
cantar (line of equal zenith distance) of 30 , the time being found by 
the equal-altitude method, invented by C. F. Gauss in 1808. 4 

Three hundred years ago, Flamsteed's double-altitude time- 
determination observations were probably accurate to about 5 
seconds of time. At the beginning of the present century, the 
accuracy expected in time determination with a conventional tran- 
sit instrument was about 1/10 second (100 milliseconds). In the 
1970s the PZT measures to 20 milliseconds on observations of a 
single star, or to 4 milliseconds on observations of, say, 30 stars - 
an exercise which can mean a lull night's work. 

in Mechanical and electrical clocks at Greenwich 

In/ Roger Stevenson 

Senior Conservation < )fflcer. National Maritime Museum 

Many ol the important advances in precision timekeeping have 
been stimulated by astronomers' demands for more accurate 
clocks. This appendix will describe the technical developments of 
the clocks used at the Royal Observatory. 

The basis of a clock 

A dock is basicallv a device that can maintain and count 
oscillations (fig. 65). The oscillator, or controller, could be a 
pendulum or balance wheel in a mechanical device, or a quartz 
crystal in iw electronically maintained clock. In each case the 
driving force is transmitted in small but regular amounts to the 
oscillator. With spring or weight-driven clocks the transfer of 
energy is achieved by >i mechanism known as the escapement, 
which, as the name implies, allows the wheels to escape and record 
the time, but at a rate dictated by the oscillator. The accuracy of a 
clock is determined by the type of oscillator used and by the 

Driving force 
e.g. falling weight, 
coiled spring, 


Transmission and division 
ol driving force 

e.g. gear-wheels and escapement, 
electronic circuits 



Time indication 
e.g. dial and hands 
digital display, 
radio signal 

balance and spring, 
tuning fork, 
quartz crystal, 
atomic resonance 


65. Clock elements. 


efficiency Wittl which the mechanism maintains the oscillator, 
whilst interfering with its motion as little as possible. Advances in 
precision timekeeping have been achieved by allowing the oscil- 
lator a progressively greater degree of control over the clock, and 
by reducing the effects from external influences such as tem- 
perature and pressure changes. 

The verge escapement 

Weight-driven clocks were introduced some time in the thirteenth 
century and these early clock movements were controlled by a 
horizontal bar-balance. The diagram shows how the balance is 
maintained in oscillation by an arrangement known as the verge 
escapement. This escapement was used in large turret clocks for 
several hundreds of years and was later adapted to control small 
66 portable clocks and watches. However, the disadvantage with the 

verge and foliot (bar-balance) design is that the balance, although 
acting as the clock's controller, does not possess a natural period of 
oscillation. Consequently, the balance is susceptible to variations 
in driving force transmitted through the wheel train,* and cannot 
complete its vibrations in equal periods of time. 

Support bracket 

Regulating weight 

Escape wheel N5§ 

66. Verge escapement 

' A train of wheels consist-, of .1 series of toothed wheels where each wheel 
ilrives its neighbour in the series. 


The pendulum clock 

A significant advance in timekeeping was made when the pendu- 
lum was applied to the simple verge clock. The pendulum pos- 
sesses a natural period of oscillation because its motion is 
controlled by the force of gravity, which is virtually constant. 

The diagram shows how the verge escapement was adapted to 
drive a pendulum. The escape wheel is now mounted vertically 
and the verge horizontally; the verge has an extension known as 
the crutch to engage and drive the pendulum. The pendulum is 
suspended by two threads or a suspension spring and the rate of 
the clock is regulated by raising or lowering the bob. 

Using the verge escapement to drive a pendulum resulted in a 
large swing of 20°-30° either side of the vertical or rest position and 
any variation in this arc gave rise to a change in rate known as 
circular error. The error increases as the pendulum swings through 
a large angle, but its effect is negligible when the arc is reduced to 
about 3 . The invention of the recoil escapement provided a means 
of maintaining a pendulum at this reduced arc. 

_ ■«- Suspension spring 


Pendulum bob 
threaded on to 
pendulum rod 


67. Verge escapement controlled by a pendulum 


The recoil escapement 

The recoil escapement has been used extensively in domestic 
clocks and has proved to be a rugged and reliable mechanism. The 
diagram shows the arrangement of the escape wheel and pallets, or 
anchor. The wheel teeth lift the impulse faces of the pallets to drive 
the pendulum rod via the crutch. The escapement is known as the 
68 recoil because the escape wheel is forced to reverse its direction for 

an instant every time the pendulum approaches the extreme of its 
swing. This occurs since the acting faces of the pallets are not 
concentric with the pallet axis. In the diagram, the right-hand 
pallet (exit pallet) has just finished receiving impulse and the pen- 
dulum is swinging to the right, but before it completes its swing, 

Pallet axis or pivoting point 


Escape wheel 

Normal rotation 
(recoil anticlockwise) 

Escape wheel axis 

- Crutch engages pendulum rod 

68. Recoil escapement. 

the left-hand pallet (entrance pallet) engages the escape wheel and 
the momentum of the pendulum forces the pallet towards the 
centre of the wheel, turning it backwards momentarily. The recoil 
escapement maintains the pendulum in a small arc but the actual 
recoil is a disadvantage, as it interferes with the pendulum's 


freedom and results in increased wear on the pallet faces. 

The large arc of swing associated with the verge escapement 
imposed a practical restriction on the length of pendulum that 
could be used, and consequently they were not normally longer 
than 6-8 inches, but with the recoil escapement it was possible to 
use a long pendulum with a relatively heavy bob. This had the 
advantage of being less easily disturbed by irregularities in the 
impulses from the clock and was also less affected by external 
influences such as draughts and vibrations. 
Flamsteed's great clocks 

The application of the pendulum to clocks, together with improved 
escapements, gave astronomers the opportunity to make accurate 
records of their observations, and pendulum-controlled clocks 
were in use at the Royal Observatory from Flamsteed's day up until 
the 1940s. The two 'Great Clocks', made by Thomas Tompion, were 
installed in the Observatory in 1676; they were unusual because 
their pendulums, which had an effective length of 13 feet and beat 
two seconds per swing, were suspended above the movements. 
This requited the crutch of each movement to be extended ver- 
tically upwards so that it would engage and drive the pendulum 
below the bob. 

In spite of these interesting innovations the clocks did not per- 
form significantly better than the common long-case clocks of 
Flamsteed's day. In the Observatory the movements and pendu- 
lums were not encased, nor were they provided with temperature 
compensation. ' There were no real advantages in using such long 
pendulums and subsequently all the Royal Observatory's standard 
clocks were fitted with pendulums beating one second, which 
means they had an effective length of 39.14 inches. This length of 
pendulum allows the seconds hand to advance in one-second 
increments on the clock dial, but, more important, it gives the 
astronomer an audible indication of each second whilst he makes 
his observation. 

The dead-beat escapement and mercurial pendulum 
The second Astronomer Royal, Edmond Halley, ordered three 
clocks from George Graham the eminent instrument and clock 
maker. Graham, who had worked with Tompion, made two very 
important contributions to astronomical clocks. The first, made 
around 1720, was a new escapement that was similar to the recoil, 
but in Graham's design the pendulum did not force the escape 
wheel to recoil and consequently it was known as the dead-beat 
escapement. The second invention, made shortly after, was a 
means of compensating for the effects of temperature changes on 
the length of a pendulum and, as described below, was achieved 




69. The movement of a pendulum clock of 1768. 
made by John Shelton, who was George 
Graham's foreman and the actual maker oi 
regulators used at Greenwich for astronomical 
purposes from 1725 to 1925. 


by using a jar of mercury as the pendulum bob. Clocks with 
Graham's escapement and mercurial pendulum became known as 
regulators and were used in observatories throughout the world for 
over 150 years. 

Impulse is imparted to the ends of the pallets by the tips of the 
wheel teeth. The sides of the pallets are concentric with the pallet 
axis and these form the dead surfaces against which the wheel teeth 
rest in between impulses, whilst the pendulum performs its supple- 
mentary arc. The supplementary arc is the additional swing of the 
pendulum beyond the minimum arc required for the pallets to 
receive impulse and for the wheel to escape. 

Pallet axis or pivoting point 

Dead surface 


Escape wheel 

■> Crutch 

71. Dead-beal escapement. 

With the influence of the escapement and circular error reduced, 
Graham recognized that changes in the rates of his clocks resulted 
primarily from changes in temperature. A clock with a common 
iron or steel pendulum rod will lose as the rod extends owing to a 
rise in temperature and gain as the temperature falls. The magni- 
tude of the error for a steel rod is about 0.5 s/day (seconds per day) 
for a temperature change of i°C, or a difference in rate between 
summer and winter of approximately 4 s/day. To overcome the 
problem, Graham had to ensure that the distance from the pendu- 





lum's suspension to its centre of oscillation (the point where the 
addition or subtraction of mass will not alter the pendulum's 
period of oscillation), situated near the centre of the bob, would 
remain constant whatever the temperature. After experimenting 
with various combinations of metal rods, in an attempt to use their 
differences in expansion to maintain the bob at a constant height, 
he finally produced the mercurial pendulum in which expansion 
and contraction of the steel rod is compensated by a rise or tall of 
the mercury contained in a glass jar. 

John Harrison's gridiron pendulum 

Strangely enough, none of the four regulators provided by 
Graham for the Royal Observatory was fitted with the mercurial 
pendulum. The first three clocks supplied between 1721 and 1725 
are described as having 'simple pendulums' (wood or steel rods). 
I larrison produced his compensation pendulum around 1728 and 
subsequently Graham up-dated at least two of his earlier regu- 
lators by replacing their simple pendulums with the gridiron type, 
whilst the fourth clock, delivered in 1750, was fitted from new with 
gridiron compensation. 

1 larrison had found by experiment that brass expands more than 
steel in the ratio of approximately 3:2 for the same temperature 
change. Therefore, to obtain sufficient compensation for a seconds 
pendulum, the lengths of the brass and steel rods will need to be 
about 9 feet and 6 feet respectively. I larrison avoided the problem 
of an excessively long pendulum by arranging the rods as shown in 
the diagram; the arrows indicate the directions of expansion for a 
rise in temperature. The rods are pinned to the horizontal cross- 
bars in a manner that allows all the steel rods to expand down- 
wards and the brass upwards, so that the cumulative effect is to 
maintain the bob at a constant distance from the suspension. 

Further improvements 

Maskelyne, the fifth Astronomer Royal, acquired two regulators 
with dead-beat escapements from the chronometer maker John 
Arnold, in 1773. Both clocks were fitted with the gridiron pendu- 
lum, one having the conventional nine-rod configuration but the 
other used only five. This was achieved by using zinc in place of 
brass for the two upward expanding members and steel for the 
remaining three rods. Zinc has a greater coefficient of expansion 
than brass and this enabled Arnold to reduce the total length of rod 
necessary for compensation. 

Arnold was also responsible for improving the earlier Graham 
regulators by fitting them with jewelled pallets to reduce friction 
and wear. The ruby jewels were inserted in the steel pallet frame 



and provided the impulse and dead surfaces. Arnold's own regu- 
lators were jewelled in this manner and it also became common 
practice in the manufacture of high class regulators to fit jewel 
bearings for the escape wheel and pallet pivots. 

Suspension spring 

, Steel rods 

I _ Brass rods 

Arrows indicate direction 
of expansion with a rise 
in temperature 

Pendulum bob 

Rating nut 

72. Gridiron pendulum. 

Electrical timekeeping 

There were no significant advances in horology during John Pond's 
time as Astronomer Royal, but the first public time signal was 
introduced in the form of the time-ball mounted above the east 
turret of the Observatory in 1833. Initially, the dropping of the ball 
was achieved manually, with the operator using a conventional 
regulator to time the event, but from 1852 the ball was dropped 
automatically by Gharles Shepherd's electric clock. Pond's suc- 
cessor, George B. Airy, saw the potential of using Shepherd's clock 
in conjunction with the telegraph network, and this electric clock 
was at the heart of Airy's time-distribution system (see ch. 3). 

Shepherd's regulator (still in full working order) uses a conven- 
tional seconds pendulum with mercurial compensation, but the 




pendulum is maintained by the fall of a small weighted lever 
known as a gravity lever. A simplification of the electromagnetic 
arrangement is shown in the diagram. When the pendulum swings 
to the left, it lifts a pivoted catch (a), which in turn releases the 
gravity lever (b) to impulse the pendulum as it returns to the right. 
Hence, the pendulum is impulsed in one direction only and the 
quantity of impulse, being determined by the force of gravity on 
the lever, remains constant. As the pendulum completes its swing 
to the right, it makes contact with (C) and the electro-magnet (D) is 
energized. The coil attracts its armature (c) and as the tail of the 
armature rises, it throws the gravity lever up so that it can be 
retained by the catch. This completes the sequence of impulsing 
and resetting. 

Pendulum rod -» 
73. Shepherd's gravity oscipcmcnt. 


The pendulum also makes contact with (A) and (B) which are 
responsible for completing the circuit with the clock driving coils 
(E) and (F). Shepherd's clock operates in reverse when compared 
to a conventional weight-driven movement, since the rocking of 
the pallets turns the escape wheel, so that the escapement drives 
the clock instead of the clock driving the escapement. In the 
original design, the polarities of the coils (E) and (F) were reversed 
at each swing of the pendulum, thus the bar magnet, attached 
across the axis of the pallets, was simultaneously attracted by one 
coil and repulsed by the other. 

Electric clocks were in their infancy when Shepherd supplied his 
regulator to the Observatory; his method of contact-making inter- 
fered with the free swing of the pendulum, but the seconds pulses 
produced by the clock could be used to operate subsidiary, or slave 
clocks, at considerable distances from Greenwich. 

Dent's regulator 

In 1871 the Observatory purchased a new Sidereal Standard Clock 
from E. J. Dent. Airy worked in close co-operation with Dent and 
this particular regulator. Dent No. 1906, has an escapement made 
to Airy's own design which is known as the chronometer dead-beat, 
as it combines features from both escapements. 2 

The regulator was made to a very high standard and it proved to 
be accurate enough to reveal the effects on its rate of barometric 
pressure changes. The effect is very small and amounts to approxi- 
mately o. 1 s/day for average changes in pressure. Airy's assistant, 
William Ellis, devised a means to compensate for these errors, by 
arranging for a fixed mercurial barometer to raise or lower a magnet 
to an extent determined by the pressure in the clock case. This 
magnet attracted two bar magnets attached to the pendulum and 
altered its period of oscillation to compensate for pressure changes. 
In later clock designs, a more effective solution to pressure changes 
was achieved by placing the movement and pendulum in a partial 
vacuum and maintaining the reduced pressure at a constant level. 

A more compact arrangement of John Arnold's zinc and steel 
gridiron pendulum was developed around 1800 by making the zinc 
in the shape of a tube, which was free to slide up or down the steel 
pendulum rod. DentN'o. 1906 was fitted with this form of tempera- 
lure compensation and several of the Royal Observatory's older 
regulators were converted to zinc and steel compensation, in place 
of their original gridiron or mercurial pendulums. 

The Riefler escapement 

Greenwich acquired its last pendulum-controlled clocks during Sir 
Frank Dyson's time as Astronomer Royal. It had become apparent 




that further improvements in timekeeping could be achieved by 
freeing the pendulum from having to perform any work. A step 
towards the free pendulum was made by Dr. Siegmund Riefler of 
Munich in 1891, with his special form of escapement that impulsed 
Ihe pendulum via its suspension spring. The clock was electrically 
maintained and therefore il was not necessary lo provide access to 
the movement for winding. Consequently, the movement and 
pendulum were mounted in an evacuated cylinder and were sub- 
stantially isolated from external effects. Under these conditions, 
the Riefler clocks attained an accuracy approaching 0.01 s/day. In 
spite of the Uieflcr's success it was not until 1922 that Greenwich 
used a clock with Riefler' s escapement to replace Dent No. 1906, 
but this regulator, made by an Englishman, E. T. Cottingham, was 
soon superseded by the Shorlt free-pendulum clock in 1925. 

The free-pendulum clock 

Shortt free-pendulum clocks were widely adopted by observa- 
tories and they have the distinction of being the first timekeepers 
capable of detecting the slight seasonal variations in the rate of the 
Earth's rotation (seech. 5). 

The Shortt clock actually consists of two clocks: a master with the 
free pendulum and a secondary or slave to maintain the free 
70 pendulum and to count its oscillations. Both clocks emplov a 

gravity lever to impulse the pendulum once every half minute. 

A pivoted hook (a) moves with the pendulum and gathers a 

74 tooth of the count wheel once every two seconds. As the wheel 
rotates, the vane (b) turns with it and every 30 seconds the vane 
trips the catch (c) and releases the gravity lever (d). A small pivoted 
roller, attached to the lever, rolls down the pendulum's impulse 
plane and, after the impulse has been given, the vertical section of 
the gravity lever contacts the armature (e). This closes the circuit; 
the coils attract the armature and the gravity lever is thrown back to 
be retained by its catch. 

In the master clock, the free pendulum is maintained in a similar 
manner to the slave, but the gravity lever is considerably lighter. 
This is because the clock works in an environment of reduced 
pressure and the pendulum does not have to operate a count 
wheel. The functions of impulsing and contact making are per- 
formed by two separate levers: the light gravity lever for impulsing 
and a heavier one which is responsible for making a positive 
electrical contact. 

The timing of the impulse to the free pendulum is controlled by 
the slave clock, but the rale, and hence the time indication of the 
slave, is synchronized by a correcting pulse from the free pen- 

75 durum. For the sake of clarity in describing the interaction between 



the two clocks, the free pendulum's impulsing and resetting 
mechanism is represented as being identical to that of the slave 

J» Suspension spring 

Count wheel 

Count wheel vane 


CHL ww 


Armature return spring 

>- Pendulum rod 
74. Freo-pendulum slave movement, 

The sequence of events is as follows: 

1 The slave gravity lever is released by the count wheel vane. 

2 The slave pendulum receives impulse and the gravity lever is 
reset; the contact also allows a pulse of current to advance the 
slave clock dial bv half a minute and to energize the release coil 

3 The free pendulum gravity lever is released from its catch. 


4 The free pendulum receives impulse and its gravity lever is reset; 

the closed contact also allows a current pulse to advance the free 

pendulum dial bv half a minute and to energize the synchronizing 

coil (B). 

The slave clock is regulated to have a small losing rate so that the 

pendulum can be synchronized by the correcting pulse from the 

master clock. Synchronization is achieved when the light spring 

(a), attached to the slave pendulum rod, is deflected as it runs into 

Master Slave 

I free pendulum) 





75. Connections between free-pendulum and slave clocks. 

the projection of armature (b), and the pendulum's time of vibra- 
tion is shortened during that particular oscillation. 

The material used for the pendulum rod in the Shortt clocks is a 
nickel steel alloy which is virtually unaffected by temperature 
changes. Invar, as the alloy is called, was first produced in 1899 by a 


Swiss physicist Dr. Charles Guillaume who worked in France. The 
alloy has a coefficient of expansion of the order of one millionth of 
an inch per degree centigrade; some specimens have a negative 
coefficient which means that they contract very slightly with a rise 
in temperature. This material is the natural choice for precision 
pendulum clocks, but it is still necessary to provide a small degree 
of compensation to allow for the effects of temperature on the 
suspension spring. 


iv Modern precision clocks 

by John PUkingtOtl, Head of Time Department, Royal Greenwich 
Observatory, Herstmonceux 

The developments which enabled pendulum clocks to provide 
progressively more uniform time scales were concerned with re- 
ducing the extent to which the swing of the master pendulum 
could be affected by variations in external conditions and friction in 
the clock mechanism. The uniformity of the scale was further 
improved by combining the results given by several clocks, each 
isolated as much as was feasible from the others; this procedure 
was made mare effective by the use of radio signals to permit the 
inclusion of results from clocks at widely separated locations. 
Progress in electronics eventually led, however, to the develop- 
ment of clocks in which the units of time interval were established 
through the use of effects unrelated to the motion ofa pendulum or 
other body under the action of the Earth's gravitational field. 

The first such clocks to surpass the performance of the Short! 
tree pendulum were based on the quartz-controlled oscillator; this 
still forms .in essential part of most of the 'atomic clocks' used by 
time services today, but in this application it has only the role ofa 
flywheel, smoothing out short-term variations of rale but itself 
subject to long-term control by more fundamental effects. The 
quart/ oscillator has been used for over fifty years in telecommuni- 
cations systems, and has recently become familiar as the heart of 
the quart/ wrist-watch in which it is usually designed to have a 
stability of a few seconds a month; stabilities several thousand 
times better than this are now attainable, at higher cost, for use in 
precision instruments. 
76, 77 Quart/, is a naturally occurring crystalline mineral which is also 

produced artificially. It is useful in timekeeping because it has 
consistent mechanical properties that are not strongly affected by 
temperature and because it displays the piezoelectric effect; this 
means that an electric field applied along certain directions within 
the crystal produces a small change in the shape of the crvstal and, 
conversely, that a change in its shape produces electric charges at 
its surface. Mechanical vibrations of the crystal produce oscillatory 
voltages between metal electrodes applied to its surfaces, and if 


these electrodes are connected in a suitable link between the input 
and output of an electronic amplifier, the amplifier can supply the 
energy required to maintain vibrations at one of the natural 
resonant frequencies of the crystal. 

The frequencies at which a particular piece of quart/ will 
resonate are determined by its size and shape and by the elastic and 
mechanical properties of the quart/.; the crystal is therefore 
accurately cut, ground, polished, and cleaned so that the wanted 
resonance occurs at a suitable frequency, and the electrode struc- 
ture and the supports for the crystal are designed to interfere as 
little as possible with the chosen mode of vibration. Operation in 
unwanted modes is suppressed by reducing the gain of the 
amplifier at the corresponding frequencies. 

In quart/ clocks elastic vibrations of the quart/ crystal therefore 
take the place of the vibrations of the pendulum, and the 
mechanical escapement and gear train which maintained the 
pendulum in oscillation and counted its beats are replaced by an 
electronic amplifier and frequency divider producing electrical 
time markers and driving a display. All the quartz clocks that 
contributed to the Greenwich time service from 1939 to 19(54 used 
resonators which operated at around 100 kHz and were designed 
at either the National Physical Laboratory or the Post Office 
Research Laboratories. In more recent precision quart/ oscillators 
the frequency is usually higher, often around 5 or 10 MHz. 













76. Quartz crystal clock. 

The dependence of the frequency of vibration on temperature- 
can be made very small by choice of theangle between the cut faces 
of the quart/ resonator and the natural faces of the crystal, and 
the small size of modern crystals makes it simpler to keep their 
temperature constant and protect them from variations of atmos- 
pheric pressure; a complete precision quartz oscillator package, 
including an oven for temperature control, occupies less than 1 litre 
and weighs less than 1 kilogram. It was more difficult to isolate 
pendulum clocks from environmental effects. 

But even modern crystals show progressive changes of rate ofa 
few microseconds per day per day which are known as ageing and 
are often associated with residual surface contamination, defects in 

77- Quartz crystal clock. This 
is the only quartz crystal 
clock still in service at the 
RGO. It incorporates a 
lenticular AT-cut crystal 
oscillating 2.5 MHz, and is 
one of a pair purchased from 
the L'SA in 1964, following 
tests made at the US Naval 


the molecular structure of the crystal, changes in the stresses 

imposed by the mountings, and exposure to radiation such as that 
from radioactive materials or cosmic rays. Moreover, the crystal 
does not provide an independent standard oi frequency or time- 
interval, since its frequency is primarily determined by its si/e but 
an easily detectable difference in frequency can be produced by a 
difference in size which is too small to be measurable. In the absence 
of other techniques, measurements of frequency until the mid 
1950s had to be related in some way to astronomical determinations 
of time. The astronomical observations needed, once the rotation 
of the Earth had been shown to be variable, were of the apparent 
motion of the Moon and the planets relative to the stars. Their 
precision is relatively low (around o. 1 second) so the observations 
must extend over many years, and the analysis is complex. It 
is fortunate thai more precise and accessible standards of time 
interval are now available, based on identical resonators which 
can be obtained in large numbers and apparently free from manu- 
facturing tolerances: indistinguishable atoms of the same kind. 

It has been known since the mid-nineteenth century that gases 
which are hot or are carrying an electric discharge emit light in 
'spectral lines' whose wavelengths are characteristic of the 
chemical elements present in the gas; perhaps the most familiar 
example is the yellow light of the sodium vapour lamp. Much of 
modern astronomy is concerned with consequences of this fact, 
and it is at the basis of the quantum theory ot atomic structure. 
According to this, an isolated atom may emit or absorb electro- 
magnetic radiation (e.g. light, radio waves) onlv by undergoing a 
transition between two more-or-less sharply defined energs states 
which are characteristic of the type ol atom. The frequency of the 
radiation is directly proportional to the difference between the 
energies of the two stales. The frequencies associated with visible 
light are too high for direct measurement at present, but fairlv 
accurate values may be calculated from the measured wavelength 
and the velocity of light. 

When an atom emits or absorbs ,1 'quantum' of visible light there 
are usually marked changes in the effective separation between the 
electrically positive nucleus and the outermost negative electrons: 

the atom changes its si/e. There are, however, less drastic ways 
in which the components ol ,m\ atom or molecule can be re- 
arranged. I he energy differences are still well defined but may be 
millions of times smaller, and the associated frequencies are the:; 
accessible through radio techniques. 

In almost all clocks currently used by time services to establish 
reference scales of atomic time the fundamental reference is the 
frequency associated with a specific hyperfine transition in the 






caesium atom of mass number 133; the definition of the SI second 
(Systeme International) is based on the same transition. Caesium is 
a soft, light, reactive metal chemically similar to sodium. The de- 
scription 'hyperlinc' is derived from the hyperfine structure that is 
observed in lines in the visible spectrum because the main energy 
levels of the atom are split into several levels of almost identical 
energy. The clocks make use of transitions between two levels in 
the split 'ground slate'. 

In this particular transition the spin of the outermost electron 
Of the caesium atom, which makes the atom behave as a weak 
magnet, 'flips over' with respect to the spin of the atomic nucleus 
while the orientation of both is controlled by a weak external 
magnetic field. The mechanism of the clock is designed tocontrola 
good quartz oscillator by establishing a known relationship be- 
tween its frequency and that of the microwave radiation which 
most effectively stimulates this transition. 

The atoms are interrogated by an atomic-beam technique which 
was developed for the investigation of atomic properties. The first 
recorded suggestion that it might be used in a standard of fre- 
quency was made by I. I. Rabi of Columbia University in a lecture 
given in 1945, iln d the first atomic-beam device to be used for 
regular frequency calibrations was brought into operation in June 
1955 by I.. Essen and ). V. L. Parry of the National Physical 


Caesium atoms are evaporated Irom metallic caesium in an oven 
within a vacuum envelope and pass through a series of slits to 
form a beam of isolated atoms in which equal proportions of atoms 
are in the two relevant energy states. A strong localized non- 
uniform magnetic field deflects atoms in the two states by differing 
amounts to form two divergent beams; these then pass through a 
resonant cavity which concentrates and confines the microwave 
energy fed to it from an amplifier. The microwave frequency of 
about 9192 Ml 1/ is synthesized from the output of a quart/ crystal 
oscillator. The cavity is immersed in a weak, uniform magnetic 
field whose axis is perpendicular to the beam. After passing 
through the cavity the atoms enter a second region of non-uniform 
field and .ire again deflected according to their magnetic state; a hot 
wire is placed to intercept only those whose stats has changed 
during passage through the cavity, and ionizes them so that they 
may be detected electronically. The frequency of the quartz oscil- 
lator is controlled by the output signal from this detector. In fact the 
control signal is generated by making small repetitive variations in 
the relationship between the frequency of the oscillator and that of 
the microwave field, and sensing the effect of these variations on 
the output from the detector; the oscillator is automatically kept 


tuned to the frequency that would maximize the proportion of 
atoms changing state in the absence of the deliberately introduced 
variations, lime markers are produced, as in the quartz clock, by 
frequency dividers and counters driven by the quartz oscillator. 



State selector 
magnet A 

State selector 
magnet B 

Transition frequency input 

Output signal to 
correct quartz oscillator 

79. Caesium beam lube. Chro mi u nu i (Bhauches SA, Switzerland), p. 4. 

In the years since 1955 standards laboratories in several countries 
have built and operated large caesium-beam devices designed to 
reproduce SI time intervals as accurately as possible by permitting 
the evaluation of all known causes of bias, while smaller instru- 
ments which are commercially available will reproduce SI intervals 
subject to undetermined but small biases of up to about 1 /is per 
day. The bias normally remains almost constant when a particular 
instrument is operated in a controlled environment and so can be 
calibrated by reference to a pre-existing time-scale or an evaluable 
standard. Instruments of this kind are used by most time services, 
including that of the RCO. to generate time-scales which agree 
among themselves to within a few microseconds per year. Inter- 
national atomic time, TAI, is based on results from instruments of 
both types. 

Frequency standards based on quantum transitions in other 
kinds of atom have also been developed; those encountered most 
often use rubidium or hydrogen. Different techniques are used in 
these devices to relate the output frequency to that associated with 
the atomic resonance, and each system has specific ad vantages: the 
rubidium gas-cell is used in commercially available devices that are 
less expensive than those based on caesium, and the hydrogen 
maser can provide better short-term stability than has been ob- 
tained with caesium; but the caesium-beam device is at present 



unsurpassed for reproducibility cind long-term stability, and it 
seems likely to continue in use as the basis of reference time-scales 
lor at least the next tew wars. 

v Time-balls in operation, 1861 

Compiled and written in his own hand by G. B. Airy, in response to a letter 
from Latimer Clark of the Electric and International Telegraph Company 
who 'had been requested for the purposes of Government to make 
inquiries as to the number of electric time-balls now in use both in this 
country and abroad'. 

List of time-balls in actual 
Name of Place 

Greenwich (Royal 

Deal (Navy Yard) 
London (E. & I. Telegraph Co. 

Office, Strand) 
London (City Observatory) 
Liverpool (E. & I. Telegraph 

Co. Office) 
Liverpool Victoria Tower 

Portsmouth (Roval Naval 

Edinburgh (Royal 


Cape of Good Hope (Simon's 

Madras (Observatory) 

Sydney (Observatory) 
Quebec (Observatory) 
Williamstown (Observatory) 
Victoria, Australia 

operation, 1861 May 

Whether dropped In/ galvanic current: 
Remarks, etc. 

Dropped by galvanic current from the 
Normal Galvanic Clock 

Dropped by galvanic current from the 
Normal Galvanic Clock at the Royal 
Observatory Greenwich 

Dropped by galvanic current from a 
clock regulated by galvanism from 
the Liverpool Observatory 

Dropped by hand 

Dropped by galvanic current 

A ball was known to exist in 1859 but no 

particulars relating to it have been found 
Dropped by galvanic current from the 

Ro\ al Observatory, Cape of Good I lope 
Believed to be dropped by hand 
A ball exists but no particulars relating to 

it are known 
Believed tobcdropped by galvanic current 
Believed to be dropped by galvanic current 
Not certain whether a ball exists at this 

place: one is believed to be in operation 


Washington, U.S. 

Known to exist only by the following 
extract from Prof. Loomis's Progress of 
Astronomy p. 289. 'At the Washington 
Observatory a ball of smaller size than 
that at Greenwich is elevated everv day 
on a flagstaff, and is lowered at the 
precise instant of twelve o'clock' 

List of places at which time-balls have been projected, 

the erection of none of which is yet known to have been proceeded with 

Name of Place In what year projected 

Hamburgh 1857 

Copenhagen 1857 

River Tyne (Shields) 1859 

Cravesend i860 

Mauritius i860 

1861 May 6 (sgd)G. B. Airy 

Source: RGO MS. 1186 Section 1 - Ci.1lv.1nic time-balls &c. 1861-6. 

vi International Meridian Conference, 
Washington, 1884 

List of delegates 

The following delegates were present: 

On behalf of Austria- Hungary: 
Baron Ignatz von Schaeffer, Envoy Extraordinary and Minister 

On behalf of Brazil: 
Dr. Luiz Cruls, Director of the Imperial Observatory ol Rio de Janeiro. 

On behalf of Colombia: 
Commodore S. R. Franklin, U.S. Navy, Superintendent U.S. Nai<al 

On behalf of Costa Rica: 
Mr. Juan Francisco Echeverria, Civil Engineer. 

On behalf of France: 
Mr. A. Lefaivre, Minister Plenipotentiary and Consul-Ceneral. 
Mr. Janssen, of the Institute. Director o! the Physical Observatory of 

On behalf of Germany: 
Baron II. von Alvensleben, Envoy Extraordinary and Minister 

On behalf of Great Britain: 
Captain Sir F. J. O. Evans, Royal Nam/. 
Prof. ). C. Adams, Director of the Cambridge Observatory. 
1 .ieut. -General Strachey, Member of the Council of India. 
Mr. Sandford Fleming, Representing the Dominion of Canada. 

On behalf of Guatemala: 
M. Miles Rock, President of the Houndary Commission . 

On behalf ol Hawaii: 
I Ion. W. D. Alexander, Surveyor-General. 
I Ion. Luther Aholo, Privy Counsellor. 

On behalf of Italy: 
Count Albert de Foresta, First Secretary of Legation, 

On behalf of Japan: 
Professor Kikuchi, Dean of the Scientific Dep'l of the University of 


On behalf of Mexico: 
Mr. Leandro Fernandez, Civil I ngineer, 
Mr. Angel Anguiano, / director of the National Observatory of Mexico. 

On behalf of Paraguay: 
Captain John Stewart, Consul-General. 

On behdlt of Russia: 
Mr. C. de Strove, Envoy Extraordinary and Minister Plenipotentiary. 
Major-General Stebnit/ki, Imperial Russian Staff. 
Mr. ). de Kotogrivoff, Conseiller d'Etat actuel. 

On behalf ot San Domingo: 
Mr. M. de J. Galvan, Envoy Extraordinary and Minister Plenipo- 

On behalf of Salvador: 
Mr. Antonio Balres, Envoy Extraordinary ami Monster Plenipo- 

On behalf of Spain: 
Mr. Juan Valera, Envoy Extraordinary and Minister Plenipotentiary. 
Mr. Emilio Ruiz del Arbol, Naval Attache to the Spanish Legation. 
Mr. Juan Pastorin, Officer of the Navy. 

On behalf of Sweden: 
Count Carl l.ewenhaupt. Envoy Extraordinary and Minister Pleni- 

On behalf of Switzerland: 
Colonel Emile l : rey. Envoy Extraordinary and Minister Plenipo- 

On behalf of the United Slates: 
Rear-Admiral C. R. P. Rodgers, U. S. Navy. 
Mr. Lewis VI. Rutherfurd. 

Mr. W. F. Allen, Secretary Railway Time Conventions. 
Commander W. T. Sampson, LI. S. Navy. 
Professoi ( leveland Abbe, U.S. Signal Offkx. 

On behalf of V enezuela: 
Seiior Dr. A. M. Soteldo, Charge d' Affaires, at the first session. 

The following delegates were not present: 

On behalf of Chile: 
Mr. Francisco Vidal Gormas, Director oj the Hydrographic Office. 

Mr. Alvaro Uianchi lupper. Assistant Director. 

On behalf of Denmark: 
Mr. Carl Steen Andersen de Bille, Minister Resident and Consul- 
General. (Did not attend any session.) 

On beha I f of Germany: 
Mr. Hinckeldeyn, Attache of the German Legation. 

On behalf of I .iberia: 
Mr. William Coppinger, Consul-General. 


On behali of the Netherlands: 
Mr. G. de Weckherlin, Envoy Extraordinary and Minister Pleni\w- 

On behalf of I'urkev: 
Rustem Effendi, Secretary of legation. 

Source. US Government, International Conference held at Washington for the 
purposeaffixing n Prime Meridian and a Universal Day, ( ktober ISS4 - Protocol •>> 

the Proceedings (Washington D.C 1884), 1 S. 



Baily, F., An Account of the Rrcd. lolm Flamsteed . . (London, 1835). 
Bigourdan, G., 'Le Jour el ses divisions. Les fuseaux horaires et 

('Association Internationale de I'Heure', Annuaire ilu Bureau des 

Longitudes (Paris, 1914). 
Bigourdan, C, 'Les services horaires de I'observatoire de Paris 

. . .', Bulletin Astronomujue, II [1921-2). 
Blair, B. E. (ed.), Time ami Frequency: Theory and Fundamentals (US 

National Bureau of Standards, May 1974). 
Brown, L. A., The Story of Maps (New York, 1951). 
Chapin, Seymour, 'A survey of the efforts to determine longitude 

at sea, 1660-1760', Navigation, 3, 7 (March 1953). 
Corliss, C. J., 77m- Day of Two Noons (Washington I1941I). 
Cotter, C. H., A History of Nautical Astronomy (London and Svdnev, 

Cotter, C. H., Studies in Maritime History, I, A history of nautical 

astronomical tables (London, 1977, microfiche). 
De Carle, D., British Time (London, 1947). 
Ditisheim, P.,etal., Pierre l.e Roy et la Chronomelre (Paris, 1940). 
Dowd, Charles N. (ed.), Charles F. Dowd. A.M., Ph.D. and Standard 

Time (New York: Knickerbocker Press, 1930). 
Ellis, William, 'Lecture on the Greenwich System of Time Signals', 

Tlte Horological journal, 1 May 1865. 
Essen, I .. The Measurement of Frequency and Time Interval (UMSO: 

London, 1973). 
l-'orbes, E. G., Tfce Birth of Navigational Science (Greenwich, 1974). 
Forbes, E. G., Greenwich Observatory, ivl. i: Origins and Early History 

(London, 1975). 
Forbes, L. G., 'the Origins of the Royal Observatory at Green- 
wich', Vistas in Astronomy, 20 (1976). 
Gazeley, W. J., Clock and Watch Escapements (London, 1973). 
Gould, R. T., 77k Marine Chronometer: its history and development 

(London, 1923). 
Guyot, E., Ilistoire de la determination des longitudes (La Chaux-de- 

Fonds, 1955). 


Guyot, E., Histoire de In determination de I'heure (La Chaux-de- 

Fonds, 1968). 
Haswell, J. E., Horology (London, 1976). 
Hope-Jones, F., Electrical Timekeeping (London, 1976). 
Howse, Derek, Greenwich Observatory, vol. Hi: Its Buildings and 

Instruments (London, 1975). 
Howse, Derek, and Hutchinson, B., 77k Clocks and Watches of 

Captain fames Cook, 1769-1969 (Antiquarian Horology reprint, 

Howse, Derek, and Hutchinson B., The Tampion clocks at Greenwich 

and the dead-beat escapement (Antiquarian Horology reprint, 1970- 1 ). 
Howse, Derek, 'Le Bureau britannique des Longitudes', 

L'Aslronomie, Oct. 1978, 413-25. 
Jespersen, J., and Fitz-Randolph, J., From Sundials to Atomic Clocks: 

understanding time and frequency, National Bureau of Standards 

Monograph 155 (Washington, DC, 1977). 
Kieve, J . L., Electric Telegraph: A Social and Economic History (Newton 

Abbot, 1973). 
Laycock, W. S., The Lost Science of John 'Longitude' Harrison (Ash- 
ford, Kent, 1976). 
Leigh-Browne, F. S., 'The International Date Line', Tlie Geo- 
graphical Magazine, April 1942. 
Lloyd, H. A., Old Clocks ( Tonbridge, Kent, 3rd edn. 1964). 
McCrea, W. H., The Royal Greenwich Observatory (London, 1975). 
Maindron, Ernest, Les Fondations de Prix a I' Academic des Science* - 

Les Laureate del Academic, 1714-1880 (Paris, 1881). 
Marguet, F., Histoire de la longitude a la merau XVIII" siecle. en France 

(Paris, 1917). 
Maunder, E. W., The Royal Observatory , Greenwich (London, 1900). 
May, W. E., A Hislon/ of Marine Navigation (Henley-on-Thames, 

May, W. E., 'How the chronometer went to sea', Antiquarian Hor- 
ology (March 1976). 
Mayall, R. Newton, The Inventor of Standard Time', Popular 

Astronomy, L, no. 4 (April 1942). 
Mercer, V., John Arnold and Sou (London, 1972). 
Michaelis, A. R., From Semaphore to Satellite (Geneva, 1965). 
Morando, B., 'Le Bureau des Longitudes', L'Astronomie, 90 (June 

Neugebauer, O., A History of Ancient Mathematical Astronomy 

(Berlin, I leidelberg, and New York, 1975). 
Pedersen, O., & Pihl, M., Early Physics and Astronomy: a historical 

introduction (1974). 
Perrin, W. G., The Prime Meridian', Mariner's Mirror, XIII, no. 2 

(April 1927). 




Quill, Humphrey, lolm Harrison, the Man who found Longitude 

(London, 1966). 
Quill, Humphrey, lolm Harrison, Copley Medallist, and the £20,000 

Longitude Prize (Antiquarian Horological Society, Monograph 

no. 1 1, 1976). 
Rawlings, A. I.., Tiie Science of Clocks and Watdus(Loudor\, aidedn. 

Sadler, D. H., Mini is not Lost: a record of two hundred years of 

astronomical navigation with the Nautical Almanac, 1767-1967 

( London, 1968). 
Sadler, D. H., 'Mean Solar Lime on the Meridian of Greenwich', 

Quarterly Journal of the Royal Astronomical Society, 19, ( 1978). 
Smith, H. M., 'International lime and frequency coordination', 

Proceedings of the IF.F.E. 60, no. 5(Mav 1972). 
Smith, H. M., The Bureau International de I'Heure', Proceedings of 

the 8th Annual PTTI Applications and Planning Meeting. Nov.- 

Dec. 1976. 
Smith, H. M., 'Greenwich time and the prime meridian'. Vistas 

in Astronomy, 20 ( 1976). 
Taylor, L. G. R., The Haven-finding Art (London, 1956). 
US Government, International Conference held at Washington for the 

purpose of fixing a Prime Meridian and a Universal Day, October 

1884 - Protocols of the Proceedings (Washington, DC, 1884). 
Waters, D. W , The Art of Navigation in England in Elizabethan and 

early Stuart times (1958, reprinted Greenwich, 1978). 
Weber, Gustavus A., I'he Naval O bserva t or y , its History, Activities 

and Organization (Baltimore, 1926). 
Wheeler, G. M., Report upon the Third International Geographical 

Congress and Exhibition at Venice. Italy. 1881 . . . (Washington, 

DC, 1H85). This has a good bibliography and a good summary to 



I till titles and sources are given only for those works not listed in tin 
Bibliography (pp. 232-4). 

Author's Preface (pp. xiv-xviii) 

1 US Government, International Conference held at Washington 
October 1884 —Protocol of 'the Proceedings, 199 2m. 

2 ibid. 199. 

Chapter 1 (pp. 1-18) 

1 Neugebauer, O., A History of Ancient Mathematical Astronomy, 

2 Stevenson, E. L. (trans. & ed.). Geography of Claudius Ptolemy 
(New York, 1932), 28. 

3 Neugebauer, op. cit. 938. 

4 Taylor, H. G. K., The Haven-finding Art, 55. 

5 Preserved in the Bibliotheque Nationale, Paris, and called the 
Carte Pisane, as it was said to have been acquired from a Pisan 

6 As for example Opus Almanach magistri fohinis de monte rcgio ad 
annos xviii explicit (elicit . Lrhardi Radolt Auguslen Vindelicorum . . . 
Seplcmbris M.cccc. Ixxxviij 1 1488). This has almanacs tor the years 
1489 to 1506 inclusive and is likely to have been the edition 
carried by Columbus. 

7 Almanach 11011a plurimis annis Venturis inseruientia: per loaniiem 
Stocfflcriiium Inslingersen & lacolvn Pflaumen Vlamenscm 
mairatissimc SUpputatia: & toti fere Europe dexlro sydere hnparlita 
I '4991- 

8 MoriSOTt.S.E., Admiral of the Ocean Sea (Boston, 1942), II. 158-9, 

9 ibid. H.400-3,406. 

10 Canovai, S., Viaggi d'Amerigo Vespucci . . . (Firenze, 1817), 57~8. 

11 Werner, J., In hoc opere haec cotinentur Nona Iranslatio primi lihri 
geographicae CI' Ptolomaei . . .(Nuremberg, 1514). 

12 Pedersen, O., & Pihl, M., Early Physics and Astronomy, 258. 

13 Werner, op. cit. sig. dv r . 

14 Apian, Peter, Cosmographicus Lilvr Petri Apiani Matliematici 
StudhseCollectus (Ingolstadt, 1524), ff. 30-1. 


15 Cemnw Frisius, Cosmographicus Liber Petri Apiani Matheuialici. 
iam de nouo integritati restitutes per Geiwiumi Phrysimn (Antwerp, 
1533), ff. xv v -xvi r . 

16 Cuningham, William, The Cosmographical Glasse. conteinyng the 
pleasant Principles of Cosmographie, Geographic, Hydrographie, or 
Navigation (London, 1559), f. 107. 

17 Gemma Frisius, Gemma Phrysius de Principiis Aslronouiiae & 
Cosmographiae . . .vsvGlobieteodemediti {Antwerp, 1530), sigs. 
D2 v -D3 r . (Translation by Philip Kay.) 

18 Gemma Frisius, De Principiis Astronomiae . . . (Antwerp, 1553), 
65. (Translation by Philip Kay.) 

19 Eden, Richard, The Decades of the Neiv Worlde . . . (London, 

1555)' f. 361- 

20 Cuningham, op. cit., f. 1 10. 

21 For example, Cervantes, Miguel de, 'The famous adventures ol 

the enchanted bark', Don Quixote, 2nd part, ch. XXIX. 

22 Cervantes, Miguel de. El Coloquio de los Perron, in Obras 
Completas (Madrid, 1944), 244. 

23 Gould, R. T., The Marine Chronometer, 12. 

24 Marguet, F., Histoiredela Longitude, 45. 

25 Brown, L. A., The Story of Maps, 209. 

26 Dreyer, J. L. E., Time, Measurement of. Encyclopaedia 
Britannica, 11th edn., 984. 

27 Kepler, Johannes, Tabulae Rttdolpliniac . . .(1627). 

28 Marguet, op. cit. 7. 

29 Guyot, E., / listoire de la determination di* longitudes, 11. 

30 Sherwood Taylor, F., 'An early satirical poem on the Royal 
Society', Notes and Records of the Royal Society, Oct. 1947, 37-46. 

31 Histoire del' Academic Royaledes Sciences, I (1668), 67-9. (There isa 
fuller account in Iluygens, Christiaan, Oeuvres . . ., XXII, 218- 
26.) (Translations by Dr Barbara I laines.) 

32 ibid. 67. 

Chapter 2 (pp. 19-44) 

1 Kenyon, J. P., The Stuarts, London, 1958 (Fontana Library edn. 
1966, 123). 

2 Be van, Bryan, Charles the Second's French Mistress (London, 
'972). 53- ' 

3 Dictionary of National Biography, XIII, 820-1. 

4 Lawson Dick, O., Aubrey's Brief Lives (Penguin English Library, 
'972)- 370-1. 

5 Plumley, N., The Royal Mathematical School within Christ's 
Hospital', Vistas in Astronomy (1976), vol. 20, pp. 51-9. 

6 Baily, F., An Account of the Rei'd. John Tlamstecd . . ., 29-31. 
Henceforth cited as Baily. 

7 PRO State Papers Domestic, Entry Book 27, f. 59. 

8 Baily, 37-8. 

9 Forbes, E. G., The Origins of the Royal Observatory at 
Greenwich', p. 48, note 17, discusses St. Pierre's identity. 

10 British I .ibrary Add. MS. Birch 4393, f. 89. 1 (olograph copy with 
signature of King and Williamson (PRO/SP44/334/27-8)! 

11 Taylor, E. G. R., 'Old Henry Bond and the Longitude', 
Mariner's Mirror, 25 (1939), 162-9. 

12 Flamsteed to Sherburne, 12 July 1682 (Baily, 125-6). 

13 Memorandum by Pell, 1675 (BL Add. MS. Birch 4393, f. 930. 

14 Flamsteed, J., Ilistoria Coelestis Britannica, III (London, 1725), 
Prolegomena, 102 (translated by Mrs E. M. Barker). 

15 Flamsteed to Sherburne, 12 July 1682 (Baily, 126). 

16 Baily, 1 12. Copies in PRO/SP29/368, ff, 299 and 44, p. 10. 

17 Baily, 37, note. 

18 Baily, 39. 

19 Memorandum by Pell, 1675 (op. cit. f. 93 v ). 

20 Forbes, E. G., The Origins of the Royal Observatory . . .', 

21 Williamson to Pell, 23 April 1675 (BL Add. MS Birch 4393. f . 97). 

22 Flamsteed to Sherburne, 12 July i682(Bailv, 126). 

23 PRO/SP Dom.44, 15 (Baily, 112). 

24 Put ten ham, George, The Arte of English Poesie (1589), ed. 
Willcock, G. D., & Warner, A. (1936), 268. 

25 Howarth, W., Greenwich: past and present (London & Green- 
wich, C. 1886), 84. 

26 A Mapp or Description of the River of Thames . . . made by lonas 
Moore Gent: . . . 1662. Pen, walercolour, gold paint on vellum. 
Museum of London on loan from PRO. 

27 SPDom. 15 (Baily. 112). 

28 Howse, Derek, Greenwich Observatory, vol. Hi: Its Buildings and 
Instruments, 5. 

29 PRO/WO-47/i9b. 

30 Howse, Derek, and Hutchinson, B., The Tompion clocks at 
Greenwich and the dead -l\-at escapement, 24. 

31 Wren to Fell, 3 December 1681 (Wren Society, V (Oxford, 1928), 

32 R. Society MSS. 243 (Fl). 

33 ibid. 

34 RGOMSS. i,f. 22. 

35 Horrox, J., Opera Postuma . . . in cake adjiciuntur Johannh 
Flamstcedii, Derbiensis. de Temporis Aequalione Diatrihi . . . 
(London, 1673). 

36 R. Society MSS. 243(H). 

37 RCO MSS. 36, f. 54. 


2 3 8 


38 I low sc, Derek, op. cit. note 30. 

39 Baity, 99. 

40 Ramsteed to Sharp, 29 March i7i6(Baily, 321). 

41 ibid. 

42 Edited translation in preparation as National Maritime Museum 
monograph. 1980. 

Chapter 3 (pp. 45-80) 

1 May, W, l ., ' Fhe l asl Voyage of Sir Clowdisley Shovel', /. 'us/. 
Maoig 13(1960), 324 •..". 

2 Lanoue, Capt. H.. memorial ol 173''. cited by Moriarty, H. A., 
in article 'Navigation', Encyclopaedia Britannka, 9th edn. (18X4), 
XVII, 258. 

3 May, op. cit. 

4 The Guardian, no. 107 (14 July 1713), 254-5. 

5 Osborn, James M., "That on Whiston" by John Gay', Biblio- 
graphical Society of America, Papers, Ivi ( 1962), 73. 

6 The Guardian, loc. cit. 255-6. 

7 Whiston, W., and Ditton, II., I Nteit' Method tor Discovering the 
Longitude both at Sea and Land (London, 1714). 

8 Osborn. op. cit. 74. 

9 Swift, Arbuthnot, Pope, and Clay, Miscellanies, the Fourth 
Volutin: Consisting 0) Versesby Dr. Swift, Dr. Arlntlltnot. Mr. Pope, 
ami Mr. Gay (London, 1747), 145-6. 

10 Rawson, C. J.. 'Parnell on Whiston', Bibliographical Society of 
America, Papers, Ivii (1963). 91 -, 1 iting Hi Add. 38157. 

1 1 House of Commons journal, 17(25 May 1714). 

12 ibid, (i 1 June 1714). 

13 Act 12 Anne cap. 15 (13 Anne cap. 14 bv modern notation). 
Quoted in lull in Quill, Humphrey, John Harrison, I he Man who 
found Longitude, -25-7. 

14 Could, The Marine Chronometer, 32-5. 

15 llobbs, William, Broadsheet dated 15 Sept. 1714*0 Flamsteed 
MSS. vol.69. £ 'fc» r 

16 Bull, Digby, Letter to Commissioners dated 29 Sept. 1714. 
Flamsteed MSS. vol. }6, f. 1 16V 

17 Brewster, Sir David, Memoirsof the Life, Writings, and Discussions 
of Sir Isaac Nmton (Edinburgh, 1855; Johnson reprint, 1965). II. 

18 Brewster. Sir David, 'On Sir Christopher Wren's Cipher, con- 
taining Three Methods ot finding the longitude'. Report of the 
I wenty ninth meeting ot the British Association . . . helil at Aberdeen 
September M59(i86o), first pagination, 34. Could obtained the 
solution independently from a Mr. Reece ol the foreign Office 
in 1936 and it is given by him in MS. in the interleaved copy ol 


his own book now in the possession of Col II. Quill. The 
original cryptogram at the Royal Society has been mislaid but a 
copy is said to be among the Portsmouth papers. It seems likelv 
rhal the Brewster \ersion may have contained some mistakes. 

19 Bennett. J. A., 'Studies in the Life and Work of Christopher 
Wren'. Phi) thesis, Cambridge Universitv, 1974, 263-5. 

20 Swilt, Jonathan. Gulliver's Travels (1726), Dent. The Children's 
Illustrated Classics ( 1952), 202. 

21 Paulson, Ronald, Hogarth'* Graphic Works (revised edn., 
London and New Haven, 1970). I. 169 70. 

22 Goldsmith, Oliver, She Stoops to Compter (i-j}), Act I. 

23 Maindron, Ernest, Fondations >/■' Prix ,i I'Academkdes Sciences, 


24 ibid. 23. 

25 ibid. 17. 

26 Marguet. V-., Ilistoireile la longitude, 85-7. 

27 Chapin, Seymour, 'A survey of the efforts to determine longi- 
tude at sea, 247. 

28 Brown, L. A., The StOryofMaps, 186-90. 

29 Phil. Trans. 37, no. 420(1731), 145-57. 

30 Dictionary of Ameriean Biography, IV, 345-6. 

; 1 Bedini, Silvio, Thinkers ami Tinkers: Early Ameriean Men of Science 
(New York, 1975), 1 18. 

32 Shepherd, A., ladles lor Correcting the Apparent Distance of the 
Moon ami a Star from the Effects of Refraction ami Parallax (London, 
1772), Preface. 

33 Halley, E., 'A Proposal of a Method for finding the Longitude at 
Sea within a Degree, or twenty Leagues', Phil. huus. 37, no. 421 

(1730, '95- 

34 Chapin, Seymour, 'A survey of the efforts to determine 
longitude at sea', 247-8. 

35 Board of Longitude MSS. vol. V, 10-11, Minutes 6 March 1756 
(RCO I lerstmonceux, henceforth cited as BL). 

36 Forbes, E. C, Greenwich Observatory, vol. i: Origins ami Early 
History. 120-1. 

37 Maskclvne. !\'., The British Marinas' Guide (London, 1763) 

38 BL 5/27(4 Aug. 1763). 

39 BL 5; 39-40 (9 Feb. 1765). 

40 ibid. 

(: \evil Maskelvne to Edmund Maskelyne, is May ■■jMi i\\l\i 
MSS. PST/76, ff. 104-6). 

42 Royal Warrant ol 4 March 1674/5. 

43 The Nautical Abnaaac and Astronomical Ephemeris tor the Year 1767 

44 For example: Gould, R. T., 77» Marine Chronometer, Quill, 



Humphrey, John Harrison, the Man who found longitude; Quill, 
Humphrey, John Harrison, Copley Mcttaliist. The best con- 
temporary account for the earlier history was: Anon, [probably 
James Short and Taylor White; see n. 54), Account of tlie 
Proceedings in Order to the Discovery of the Longitude at Sea 
(London, 1763). 

45 BL5/12V. 

46 Act 3 Geo. Ill, cap. 14. 

47 BL5/30V. 

48 Act 5 Geo. Ill, cap. 20. 

49 Howse, D., 'Giptain Cook's marine timekeepers'. Antiquarian 
Horology (1969), 190-205, reprinted as The Clocks and Watches of 
Captain lames Cook, 1769-1969. 

50 Beaglehole, J. C, The faumals of Captain lames Cwk, II (London, 
1961), 692. 

51 Wales, W., & Bayly, W., The Original Observations made . . . in 
the years 1772, 1773, 1774 and 1775 . . . (London, 1777), 280. 

52 May, W. E., 'How the chronometer went to sea', 638-63. 

53 Nivernois to Praslin, 21 March 1763. The subsequent story is 
told in the following manuscripts of the Academic des Sciences, 
Paris: Praslin to Ghoiseul, 28 March 1763; Choiseul to 
Academic, 31 March 1763; Academic to Choiseul, 4 April 1763; 
extract from Academic register, 16 April 1763. 

54 Chapin, Seymour, 'Lalande and the longitude: a little-known 
London voyage of 1763', Notes and Records of the Royal Society, 32 

53 Camus to Morton, 2 June 1763; Morton to Camus, 3 June 1763; 
contemporary copies among the papers of James Stuart 
Mackenzie, Mount Stuart, Rothesay, Bute, quoted by per- 
mission of the Marquess of Bute. 

56 Le Roy, Pierre, Expose succinct des TravauxdcMM. Harrison el l.e 
Roy dans la Recherche des Longitudes en Mer & des cprcuves faites de 
lews Ouvrages (Paris, 1768), 34-5. 

57 Ditisheim, P., ct al., Pierre Le Roy et la Chronometre, ioo-i, 
quoting from Berthoud, 1 "raite des Monlres a I ongitude . . . (Paris, 
1792), and Berthoud to Minister of Marine, 26 Dec. 1765 (Bibl. 
Nat.; Nouv. acq. franca is 9849). 

58 Maindron, F.rnest, Les Tondationsde Prix . . .,21. 

59 Marguet, F., Histoire de la longitude, and Guyot, E., Histoiredela 
determination des longitudes, both give excellent reviews of the 
early history of the French chronometer makers. 

60 Gould, R. T., The Marine Chronometer, 86. 

61 Lob, Decrets, Ordonnances et Decisions concernant le Bureau des 
Longitudes (Paris, 1909), 1-15. 

62 Morando, B., 'Le Bureau des Longitudes', 279-94. 


63 Anon., The Time-Ball of St. Helena', Nautical Magazine IV 
(1835), 658. 

64 Nautical Magazine (London, 28 Oct. 1833), 680. 

65 See: 'Neptune', 'The lime-ball at Greenwich', Nautical 
Magazine IV (1835), 584-6; Laurie, P. S., 'The Greenwich Time- 
ball', The Observatory, 78, no. 904 (June 1958), 113-15; Howse, 
Derek, Greenwich Observatory, ml. Hi: Its Buildings and Instru- 
ments, 134-6. 

Chapter 4 (pp. 81-115) 

1 An excellent historical description of how the day was divided 
by various peoples from the earliest times is contained in 
Bigourdan, G., 'l.e Jour et ses divisions . . .', B1-B40. See also 
Neugebauer, O., A Histon/ of Ancient Mathematical Astronomy. 

2 Bigourdan, op. tit. B8-9. 

3 Joyce, H., The Histon/ of the Post Office (1893), 283. 

4 Airy, G. B., Report of the Astronomer Royal to the Board of Visitors 
(1857), 15. Henceforward cited as Re/mrl . . . 

5 De Carle, D., British Time, 107-9. 

6 Illustrated London News, 14 May 1842, 16. 

7 PRO/RAIL 1005/235, f. 58, 3 Nov. 1840. 

8 The text of this petition does not seem to have survived (House 
of Lords Record Office). 

9 PRO/RAIL 1008/95. 

10 Booth, Henry, Uniformity of Time, considered especially in reference 
to Railway Transit and the Operations of the Electric Telegraph 
(London and Liverpool, 1847), 4. 

11 ibid. 16. 

12 Smith, H. M., 'Greenwich time and the prime meridian', 221 . 

13 PRO/RAIL 1007/393 (for I.&NWR) and prisate communication 
from Prof. J. Simmons (forGiledonian R.). 

14 Illustrated London News, 13, 23 Dec. 1848,387. 

15 Bagwell, P., The Transport Revolution from 1770 (London, 1970). 

16 Airy, Report (1849), ,6 - 
-ij RGO, MSS. 1181. 

18 KieveJ. 1.., E.lectricTelegraph-.ASocialandEconomicllistory, 104. 

19 RGO, MSS. vols. 1181 and 1182 for correspondence with SER, 
ETC, and Admiralty; vol. 1299/37 an d '300/49 for corre- 
spondence with Shepherd. 

20 Airy's journal (RGO MS. 621). 

21 The Times, 23 Aug. 1852. 

22 Ellis, William, 'Lecture on the Greenwich System of Time 
Signals', 98-9. 

23 Varley, C. F., Description of the Chronopher. . . . Pamphlet 
published by the Electricand International Telegraph Company 



about 1864 (GEO Post Office Records - Post 81 /46). 

24 The Times, 19 Juno 1852. 

25 dark to Airy, 28 Aug. 1852, RGOMSS. 1182/2. 

26 RGOMSS. 1182/ 1. 

27 Cooper, B. K., and Lee, C. K., 'Standard or /one Time', 77k 
Rn Hum/ Magazine, Sept. 1935, 159. 

28 RGOMSS. 1181/3. 

29 Baldock to Admiralty, 13 April 1852, RGO MS. 1190.01. 

30 Unless otherwise stated, sources from this table are from (11) 
Ellis, W., op. cit. (n. 22), April-July 1865, 85-91, 97-102, 109-14, 
121-4; ('") Ellis, W., 'Time signalling: a retrospect', The 
llorological Journal, Oct. 191 1, 21-3; (c) Airv, Rqvrt ( 1868), 22. 

31 Airy. K.7>c>rMi868), 22. 

32 Airy, G. B., quoted in Ellis, lecture on the Greenwich System 
of Time Signals', 123. 

33 Airy, Report (1874), 16-17. 

34 GPO/Post3o/E 9195/ 1888, file 2. 

35 Shenlon, Rita, private communication, 21 March 1978. 

36 Jagger, Cedric, Philip Paul Barraud (London, 1968), 68-9. 

37 Davies, Alun C., 'Greenwich and Standard Time', History Today, 
28(3) (March 1978), 198. 

38 R|usselll, W. J., Abraham Toilet! Osier, 1808-1903 1 1904]. 

39 The Times, 12 Jan. 1850. 

40 Smith, op. fit. 220. 

41 Edinburgh Town Council, Minutes, 4 Jan. 1848. 

42 illustrated London News, 12, 12 Feb. 1848,89. 

43 Anon., 'Greenwich time', Blackwood's Edinburgh Magazine, 63 
(March 1848), 354-61. 

44 For example; F|rodsham|, C|harles|, Greenwich time: the uiti- 
i'crsal standard of time throughout Great Britain (London, 1848), 11 


45 Anon,, 'Railway-time aggression', Chambers Edinburgh Journal, 
XV, no. 390, new series (21 June 1851), 392-5. 

46 The Times. 2 Oct. 1851, 7. 

47 ibid. 7OCL 1851,3. 

48 Illustrated London News, J Jan, 1852, 10. 

49 The Times, 21 Nov. 1851, 3. 

50 ibid. 17 Nov. 1851, 7. 

51 Nlorthcote to Airy, 10 Aug. 1852 (RGO MS. 1168/116-17). 

52 Airy to Northcote. 11 Aug. 1852 (RGO MS. 1168/1 18-21). 

53 Airy, Report (1853), 8. 

54 77« Western luminary, 31 Aug. 1852 (RGO MS. 1 181/132). 

55 Tucker to Airy, 29 Oct. 1852 (RGO MS. 1 168/133). 

56 Latimer, )., The Annals of Bristol in the Nineteenth Century 
(Bristol, 1887) and personal communication from G. Langley, 

NO 1 1 S 

County Reference Librarian, 13 Feb. 1978. 

57 Plymouth, Diivnport and Stonehouse Herald, 18 Sept. 1852,8. 

58 The English Reports, CLVII, Exchequer Division, XIII (1916), 719. 

59 ibid. 

60 ibid. 

61 77k rimes, 14 May 1880. 

62 Act 43 & 44 Vict. c.9. Unfortunately, no copy of the Com- 
mittee's report, nor a record of any debate in either House, 
seems to have survived (Mouse of Lords Record Office). 

63 Baker, E. C, 'Post Office clocks', Post Office Telecommunications 
Journal, Feb. 1954, 54-5. 

Chapter 5 (pp. 116-51) 

1 Forbes, E. G., Greenwich Observatory, vol. i: Origins and Early 
History, 150. 

2 Airy, Reports (1855), 11; (1863), Appendix III, 19. 

3 Airy, Report (1867), 2 "- and RGO MS. 1 187-01. 

4 Airy, Report (1867), 21. 

5 Bigourdan, G., 'Le jour etses divisions. . .',843,411. 

6 Corliss, C. J., The Day of Tieo Noons, 3. 

7 Weber, Gustavus A., The Naval Oliservatory, 27-8. 

S Langley, S. P., 'On the Allegheny System of Electric Time 
Signals', The American Journal of Science and Arts ( 1872), 377-86. 

9 Carson, Mrs. Ruth, private communication dated n Nov. 1968, 
quoting Dowd, Charles N. (ed.), Charles T. Dowd. 

10 Dowd, op. cit. Plate VI. 

1 1 (fl) Smith, H. M., 'Greenwich time and the prime meridian', 
222-3, where many of the primary sources are cited; (/>) Dow d, 
op. cit.; (c) Mayall, R. Newton, ' The Inventorof Standard Time'. 

12 Neio York Herald, Sunday 18 Nov. 1883, l0 - 

13 ibid. 19 Nov. 1883,6. 

14 Harper's Weekly (New York), 29 Dec. 1883, 843. 

15 Popular Astronomy, Jan 1901. 

16 Detroit News, 26 Sept. 1938. 

17 Rubio, Jose Pulido, El Piloto Mayor 
Contratacidn (Seville, 1950), 438-41. 

18 Perrin, W. G., 'The Prime Meridian', 1 18. 

19 ibid. 1 19. 

20 Anon., 'Remarques sur les Observations astronomiques faites 
aux Canaries en 1724 parle P. Feuillee, Minime', Mem. del'Acad. 
Royalcdes Sc. de Paris (1742), 350-3. 

21 Struve, Otto, 'The Resolutions of the Washington Meridian 
Conference', translation from German in Fleming, S., Universal 
or Cosmic Time (Toronto, 1885), ^5 (' n RGO Tracts, Geodesv, 


de la Casa de la 


22 Comptes-ttndus des Congres des Sciences Giographiques, Cosmo- 
graphiques, ci Commercials tenu a Aimers du U au 22 Aoi'ii 1871 
(Anvers, 1882), 11. 254-5. 

23 Borsari, Ferdinando, // Meridmno iniziale e Vara universale 
(Napoli, 1883), 61. 

24 Smith, op. cit. 222. 

25 Fleming, S., Uniform non-local lime (Terrestrial Time) (Ottawa 
l'8 7 6|). 

26 Mayall, R. Newton, 'The Inventor of Standard Time'. 

27 Fleming S., Time-reckoning and Ihe selection of a prime meridian to 
be common to all nations (Toronto, 1879). 

28 Airy to Colonial Secretary, 18 June 1879, quoted in Fleming, S., 
Universal or Cosmic Time (Toronto, 1885), 33. 

29 Colonial Secretary to Governor General, Ginada, 15 Oct. 1879, 
in Fleming, S., Universal or Cosmic Time, 31. 

30 Pia/zi Smyth to Colonial Secretary, 5 Sept. 1879, in Fleming, S., 
Universal or Cosmic Time, 35-8. 

31 'Memorandum of the Royal Society of Canada on the Unifi- 
cation of Time at Sea', Trans. R. Soc. Canada (1896-7), II. 28. 

32 De Beaumont, 11. Bouthillier, Clioix a" tin meridien initial unique 
(Geneva, 1880). 

33 Wheeler, G. M., Report upon the Third International Geographical 
Congress . . . 1881 . . . (Washington, 1885), 28-9. 

34 Smith, op. cit. 224-5. 

35 International Conference held at Washington for the pur/H'se of fixing a 
Prime Meridian and a Universal Day, October 1884 - Protocols of the 
Proceedings. Unless otherwise cited, the source of all subsequent 
information on the Washington Conference comes from this 

36 H[inks], A. R., 'Nautical rime and civil date'. Geographical 
journal, LXXXVI, 2 (1935), 153-7. 

37 None, J. W., A Neiv and Complete Epitome of Practical Navigation 
. . . (lothedn. 1831), 313. 

Chapter 6 (pp. 152-71) 

1 Pasquier, E., 'Unification of Time', quoted in journal of the 

British Astronomical Association, Nov. 1891, 107. 
2. Bigourdan, G., 'Le jour etses divisions. . .', B60-8. 

3 Decret no. 78-855 du 9 aout 1978 relatif a l'heure legale 
francaise, journal Officiel de la Refwblique Francaise, 19 Aug. 
1978, 3080. 

4 Bigourdan, op. cit. B 35. 

5 'Memorandum of the Royal Soc. of Canada . . .' (ch. 5, no. 30), 
loc. cit. 15. 

6 ibid. 48. 


7 Bigourdan, op. cit. B 36. 

8 Minutes of Conference on Time-keeping at Sea, London, June- 
July 1917 (MOD, Hydrographic Dept. MSS.). 

9 Sadler, D. H., 'Mean Solar Time on the Meridian of 
Greenwich', 290-309. 

10 Pigafetta, Antonio, 'Diary', quoted in Stanley, Lord, of 
Alderley (ed.), The First Voyage round the World (London, 
Hakluyt Society, 1874), 161. 

1 1 Dampier, W., A Neiv Voyage round the World (London: Argonaut 
Press, 1927), 255-6. 

12 Leigh-Browne, F. S., 'The International Date Line', 305-6. 

13 Hellweg, J. F., 'United States Navy time service', Pub. Astr. Soc. 
Pacific, 52, 305 (Feb. 1940). 

14 Bigourdan, G., 'Les services horaires de I'observatoire de Paris 
. . .', 30. 

15 ibid. 32-3. 

16 Bureau des Longitudes, Conference Internationale de l'heure 
. . . (Paris, 1912), Di. 

17 Smith, H. M., 'The Bureau International de l'l leure', 29. 

18 Willett, William, The Waste of Daylight (1907), quoted in full by 
De Carle, D., British Time, 152-7. 

19 Wilson, M., Ninth Astronomer Royal (Cambridge, 1951), 201. 

20 Bigourdan, G.,'Le jour etses divisions. . .',672-3. 

21 Esclancon E., 'La Distribution telephonique de l'heure et 
l'horologe parlante de I'observatoire de Paris', Annuaire du 
Bureau des Longitudes pour 1934, c. 6-11. 

22 Ordnance Survey, History of the Retriangulation of Great Britain 
(1967), 92-101. 

Chapter 7 (pp. 172-90) 

1 Smith, 1 1. M., 'The determination of time and frequency', PrOC. 
/EE, 98, II, 62 (April 1951), 147. 

2 Spencer Jones, Sir Harold, 'The Earth as a Timekeeper', Proc. R 
lust. GB, xxxiv, 157(1950), 553. 

3 Trans, int. astr. Un., x (1960), 489. 

4 'Proccs-verbaux des Seances' in Comite International des Poids el 
Mesures, 20 scrie. Tome xxv, session de 1956 (Paris, 1957), p. 
77. (Author's translation.) 

5 Blair, B. E. (ed.), Time and Frequency: Theory and Fundamentals 


6 Comptes-reiidus des Seances de la Treizicmc Conference Generate des 

Poids et Mesures (Paris, 1968), Resolution 1, p. 103. (Translation 
in Blair, op. cit., p. 11.) 

7 Smith, II. M., 'International time and frequency coordination', 



8 Comptes-rendus des Stances it In Qttinzicme Conference Generate ties 
Poiifcel Mesures (Paris, 1976). Resolution 5, p. 104. 

9 Sadler, D. H.. 'Mean Solar Time on the Meridian of 
Greenwich', 290 309: this gives an excellent account of many of 
the events related in this chapter. 

10 Decret no. 78-855 du gaoiit 1978, foe. cit. 3080. 

Appendix 1 (pp. 192-8) 

1 Edwards, Clinton R., 'Mapping by questionnaire: an early 
Spanish attempt to determine New World geographical 
positions', lina^o Minidi. XXIII (1969)21-2. 

Appendix II (pp. 199-204) 

1 All the instruments mentioned in this Appendix are described 
and illustrated in I lowse, Derek, Greenwich Observatory . vol. Hi: 
li< Buildings and Instruments (London, 1975). 

2 Perfect, D. S., 'The I'ZTof the Royal Greenwich Observatory', 
Q.\'.RAS( 1959), 223-33. 

3 Danjon, A., 'I. 'Astrolabe Impersonnel de I'observatoire de 
Paris', Hull. Astrvn., XVIII (1954), 251.' 

4 divot, L., I liMotrc de la determination de I'ltcurc, 117-19. 

Appendix III (pp. 205-19) 

1 I lowse. Derek, and Hutchinson, B., The Tampion clocks at 
Greenwich and the dead-beat escapement. 

2 Gazeley, W. J., Clock and Watch Escapements. 


i"i/v'i numbers in italic* refer to illustration* 

Abbe, Prof. Cleveland. 140. 230 

Abinger observatory and time service. 174. 

Adams, Prof. J. C. 148, 175, 229 
Addison. Joseph, 47-8, 51 
Admiralty. British, 75. 79. 91 
^BUfKffUfOH, MMS, 117 
Aholo. Hon. Luther, 229 
Airy, George B., 84. 88, 89-94, 99, 100. 1(0, 

108, 112. 117-18, 134, 136, 173. 204, 213; set 
also transit circle 

Alaska, date kept. 162 
Alexander VI, Pope, 127 
Alexander. Hon. W. D.. 229 
Allegheny Observatory. 121 
Allen, W.' F., 144-5, 230 

almanac, see GoHM&tMtf •!••< Tempt. Nautical 

Almanac, Regiomontanus 
Altona Observatory, 116 
Amsterdam mean time. 156 
Anderson, Capt., 118 
Anguiano, Angel, 230 
Anne, Queen, 41-2. 47, 52 
Anson, Admiral Lord. 63 
Apian, Peter, 7,8 
Apollonios, 1 

apparent solar time. 67. 82. 187, 199 
Arago. D. F. J..82 
Arbela, Battle of. 3 
Arbuthnot, Dr, 49 
Arnold & Son, John, 71. 72, 77. 86. 116. 212- 

Arsandeaux, chronometer maker, 77 
Arthur. President, 139 
Association, HMS.45.46 
astrolabe, prismatic, 203. 204 
Astronomer Royal. set Flamsteed, Hallev. 

Bradley, Bliss, Maskelyne, Pond, Airy, 

Christie, Dvson. Spencer Jones 
Atkinson, DrR. D' E., 171 
Atlantic Neptune, atlas, 129 
Atlantic Telegraph Co.. 117 
Atlas GnaVsMf, 44 
Atomic rime. International (TAI). 182-6. 

Aubrey, John. 21 
AurortL, frigate, 75 

Auwers. astronomer, 156 
Au/out, Adrian. 15. 16 
Azores, see meridians 

backstaff, 58 

Bain. Alexander, 83 

Baldock, Capt. Thomas, R\, 99 

Barraud and l.unds, Messrs. 104 

Barrow, John. 80 

Babes, Antonio. 230 

Baudin. Capt. N., 77 

BBC, 169 

Belhaven, Lord. 47 

Belville, John Henrv, 84-6; Mmo. 84; Ruth, 

Ben Gerson, Levi, 6 
Bering Strait. Me mendians 
Bernoulli, Daniel, 57 
Berthoud, Ferdinand. 73-8; Louis. 77 
Besant, coach proprietor, 83 
Biesta, chronometer maker, 77 
Big Ben, serdoefcs, Westminster 
Bigourdan. Guillaume, 166, 232 
Bird, John, 60 
Birmingham Philosophical Institute. 87. 105 

BtadaeoaB"s Magazine, 106 

Blaeu, Willem.'l28 

Bliss, Nathaniel, 64 

Bond. Henrv, 25, 27 

Booth. Henry, 87-8 

Borda. Jean-Charles. 77. 78 

Bouganville, Louis Antoine de, 78 

Bouguer, Pierre. SB 

Bovle. I Ion. Robert. 16 

Bradley. James. 63, 64; Ml al~o transit 

BradshaafS Railioau Guitlc, 88 
Brahe, Tycho. 13 
Branly. tdouard. 163 
Breguct, Abraham Louis, 77; Louis. 77 

Bristol Times, 112 

British Broadcasting Company (later 

Corporation), 169 
British Horological Institute. 94. 99 
British Summer Time (BST). 167 
Brouncker. Lord, 24. 33 
Brunei, Isambard Kingdom, 118 



l\l)i \ 


Bruyeres, H. P.. 87 

Buachc. P., 78 

Buckingham, Duke of. 19 

Bull, Digby. 53 

Bureau dcs Longitudes, see Paris 

Bureau de I'Heure (BUI). 

164-6. 182-6 
Burslow, Surrey, 41 
Bute. 3rd Earl of, 73 
Button's Coffee Mouse. 48 

caesium time standard. see clocks, atomic 

Cambridge, Jesus College, 23 

Cameron. Dr, 114 

Campbell. Capt. John, RN, 60, 63 

Camus, Charles-Eticnne, 73-5 

Canary Is., see meridians 

Cape Verde Is., see meridians 

Caroche, Noel-Simon, 78-9 

Carte Phme, 3, 235 

Carter, George, hatter, 104 

Chi ieContmacOn, 127 

Cassini. Giovanni Domenico. 15, 16 
Ca5Sini de Thurv. Jacques-Dominique. 62. 

Castlemaine. l-adv. Duchess of Cleveland, 

Catharine of Braganza. Queen. 19, 21 
Central European Time, 156 
Central Telegraph Office. London. 94. 95. 96. 

100 1 

Centurion, I IMS, 68 

Cervantes, Miguel de, 10 

Chambers Edinburgh fcwnof, 107 

Channell, Baron of the Exchequer, 1 14 

Charles II. 15, 19, 23 32. 40-1, 44, 61. 66. 69 

charts, sea, 3 

Chatham, vachl, 58 

Chelsea College. 23, 25, 28 

Chicheley, Sir Thomas, 30. 32 

Christ's, 23.47 

Christie, W.H.M., 136 

Chronograph. Airy's. 100 

early, 14 

marine, in Britain, 67-72; in France, 73-8 
tee sbo longitude and individual motors 

Chronopher. 95, 96, 101 

'Chronos', 108-9, 112 

C i,l illcm. 153 
circles, reflecting, 60 
civil time. 81, 149 
Clairaut, Alexis-Claude, 62 
Clark. Edwin. 96. 99 
clepsydra, 10 

accuracy of, 181 

atomic, 180-2, 187-90. 222. 223-6. 225 

caesium, xv atomic 

electric, 84, 89-92. 213-15 

Hamsteed's vear-clocks. 33-40. 39, 172. 

for astronomical purposes. 13 
free-pendulum, 172-4, 210, 216-19. 217, 

Greenwich gate dock, 92. 93 
invention of, 3 
'mean clock'. 173, 182, 184, 190 

Mean Solar standard. Greenwich, 90- 1. 93. 

94, 173. 213-15 
pendulum, 13, 14, 207 
quartz, 173-5, 181,220-3,221.222 
Royal Exchange. London, 91 
sea clock, see Chronometers 
Shepherd's electric, 89-92. 213-15 
Sidereal Standard. Greenwich, 100, 173, 

Westminster ('Big Ben'), 91, 96. 100-1. 102, 


see id*' nnln-idiiiil makers and d es i a n ei s 

Clockmakers, Worshipful Company of. 86 
Colbert, jcan-Baptiste. 15. 16 
Collier. W. F„ 113 
Columbus, Christopher. 4, 5 
on the prime meridian ( 1634). 128 
1st International Geographical Congress 

(1871). 131, 2nd (1875). 131; 3rd (1881), 

7th International Geodesic Conference 

(1883), 136-8 
US General Railway Time Convention, 

International Mendian Conference (1884), 

xiv. 136-7. 138-51, 178 
International Telegraph Congress (1884), 

International Time Conferences (1912. 

1913), 165-6 
Anglo-French Conference on Timekeeping 

at Sea (1917), 79, 157. 167 
Inter-Allies Conferences on Scientific 

Academies (1918, 1919), 166 
International Astronomical Union (1928). 

General Conferences on Weights and 

Measures (1967. 1975). 182. 186 

Ctmnajssancedes Tempt, 66, 73. 78-9, 156 

Cook. Capt. James. RN, 71. 149, 197 

Copley Medal, 68 

Coppinger, William, 230 

Cosmic (Cosmopolitan) Time. 132, 144, 152 

Coltingham, F.. T.. 216 

Courtanvaux, Marquis de. 75 

Coventry, Secretary, 26 

Cox, Sir Richard, 50 

Crabtree. William, 23 

cross-staff, 6, 7 

Croslhwait, Joseph. 44 

Cruls.DrC, 139,229 

Cuningham, Dr William, 6, 10. II 

Curtis I'. March, 113-14 

Da Gama, Vasco. 4 

D'Alembert, Jean. 62 

Dampier, William, 161 

Danjon. Andre, 203. 204 

Dale Line. International, 132. 160-3 

Day, astronomical. 137. 149. 149-50. 156-60; 

changes in length of. see Earth; civil. 145, 

149. 156 60; division of. XV, 81; mean solar. 

145; nautical, 149, 149-50; unification of 

civil and astronomical, 149, 156-60; 

universal. 137. 144-8. 150-1. 186 
Daylight Saving Time. 166-7 
de Beaumont. Prof.. 135 6 
deBille.C.S. A.. 230 
de Carle, Donald. 86. 232 
decimalization of angles and time. 136, 137, 

de Foresta, Count Albert. 229 
de la Crenne, Verdun, 77 
de la I lire, Philippe. 15 
Delambre, Jean-B.iptiste-Joseph. 78 
del Arbol. Emilio Ruiz. 230 
de la Rue, Warren. 101 
Dent, E. J., 96. 105; Regulator No. 1906, 172. 

Dentrecasteaux, J. A. B.. 77 
Depffari, I IMS, 69 
De Ruvter. Admiral, 19 
DesBa'rres. J. F. W.. 129 
de Struve. C, 230 
Dtvillo, Deputy, 153 
dip, magnetic, 25 
discovery, age of, 4 
Disraeli, Benjamin. 112 
Ditton, I lumphrey. 47-50, 56 
Double British Summer Time (DBS I). 167 
Dowd. Charles F„ 121-7, 122, 132, 152, 232 
Dudley, Robert, Earl of Leicester. 31 
du Quesne. Lt.-Gen., 16 
Dye, Dr, 174 
Dyson, F. W., 165, 169-70, 215 

Efljjte, HMS, 45 

Earnshaw, Thomas, 72 

Earth's rotation rate, changes in, 33-40. 

175-7. 176. 182, 187 
East India Company (British), 64, 68, 72. 80. 

Eaton. II ., 101 
Echeverrier, J. F., 229 
Eden. Richard, 10 
Edinburgh, Royal Observatory. 173-5. 202; 

Strata time-balls, time-guns 
Eiffel Tower. set time signals 
electric telegraph. 83—1; xeeba Longitude 
Electric Telegraph Co. (ETC). 90, 92. 96-7, 

Electric and International Telegraph Co. 

(E&ITC).94 6. 100-1 
Elizabeth I. 31 
Ellis, Henry. 109.110 

Ellis, William, 215. 232 
1 'mi'uiv. L'. frigate, 77 
Ephemerides. tee Nauliad Almanac, 

Ephemens rime (ET), 83. 176, 178-80 
Equation ol Time. 37-40. 3S. 67. 82, 199 
Equinoctial Time, 83 
I'rvington. Mr, 1 14 
e scape m ents, 205; anchor (or recoil). 208-9. 

208; dead-beat, 172, 209 12. 210, 211; 

verge. 206, 206, 207 
Essen. Dr L.. 174. 180. 224. 232 
Euler, 1.62,71 

Evans, Capt Sir Frederick J. O., RN, 140. 229 
Evelyn, lohn. 16 

Fell, Bishop, 32 
Fernandez, I... 230 
Ferro. tee meridians 
Feuillee, Abbe Louis, 129 
Field, Cyrus W. 117 
Fiji Is., date kept, 162 

Firebrand. HMS, 45 

Flamock. standard-bearer. 31 

Fl.unsteed, |ohn. 21 44. 39, 43. 50, 53, 61. 

187, 209; set alv Alia- Cbeksft, clocks. 

Hifloria Coekstb 
Flamsleed, Margaret, 40 
Flamsteed, William. 41 
Flamsteed 1 louse, 92 
Fleming, Sandfard, 126, 132, 133, 141 2. 144. 

Fleuneu. Charles P. C..77 
Foerster, Prof. W.. 165 
Fortunate Isles, see meridians 
France, survey of, 16. 17. 25 
Franklin, Benjamin, 59 
Franklin. Cdre. S. R.. USN, 229 

French, J., 96, 100 
Prey, Col. Emilc. 230 

Freylinghuysen, I Ion. F. T.. 138 

Galileo Galilei, 12, 13, 16, 54. 197 


galvanism. 84. 89. 91 

Garner, J. H., 115 

Gascoigne, William. 21. 23 

Gauss. C. F., 200, 204 

Gay, lohn, 49-50 

Gemma Frisius, 6, 9 

George 1, 42 

George III, 64 

George of Denmark. Prince. 41 

Gloucester. I lumphrey. Duke of. 31 

Godfrey. Thomas, 59 

Goldsmith, Oliver, She Stoops to Conam r. 56 


Gould, DrB. A.. 118 

Gould, Cdr. Rupert, RN. 12, 53, 77, 232 

Graham. George. 68, 172-3, 210, 212 

Graphic Til,; 93 

Cn-at Eastern. SS. 117-18. 119 

2 5 


Great Exhibition (1851), 90. 105-6 Pyramid, st meridians 

( ..-tic. 29, 31 

Greenwich Civil Time (CCT). 160 
Greenwich Mean Time (GMT), \iv, 72, 

151. 153 6. 157-60, 185-6, 190, 204 
King- House, 34-5 
Queen's House. 32, 34 5 
Royal observatorv. 93, 143, 188, IS9, 

Board of Visitors. 41-2. 89: building of. 

31-3; Flamsteed House. 92; foundation 

of, 23. 27-30; Great Room (Octagon 

Room), 32. 36. 39. 40. 200; longitude 

operations. 116-18 
Station. 94, 1(X). 101 
lime, lor astronomers, ch. 2; for Great 

Britain, ch. 4; for navigators, ch. 3; for 

the world, ch. 6 
Greenwood, W. M., 157 
Gregoirc. Citizen, 78 9 
Gregorian Calendar. 162 
( Ire-sham College, 15 
Cmmlian. Tin; 47 
Guillaumc, Dr Charles, 219 

Gumbo's Travels, 56 

Gunning. Bishop, 28 
Cwvn. Nell, 21 

Gytden,Prof., 151. 156 

I l.idlev, John, 58-9, 62 
Hall, Prof, Asaph, 165 
Hall, Capl. Basil, RN. 86. 87 
llallev. Fdmond. 32. 41-4, 50, 58-9, 61-2, 
68. 175. 200 

Harper's WetUy, 126 

Harrison, John, 64. 67-72. 73 5. 78, 212; HI. 
68. 70. 73; 1 12. 68. 73, H3. 68. 73; 1 14. 68 9. 
70. 73-1. 75, 78; 1 15. 71 -2 

I l.irrison, William, 69 

Harvard Observatorv. 116-17. 118 

I lenrv V, 31 

Henry VIII, 31 

1 tally, |ohn. Me Bclville 

I lenrv the Navigator of Portugal. Prince. 4 

I lerbcrt, Mr, 92 

Hcrschel. Sir John. 83, 116 

Hcrstmonccux. Roval Greenwich Observ- 
atory, 175, US, 202. 225 

hemta la fnutque, heart i In torque, 148 

I lill. Mr, 27 
Hill. Rowland, 86 

I Unckeldeyn, 1 tar, 230 

Hipparcho's, xvi. 2. 12. 127. 148. 149. 192-4 

HinmieHe. I.', corvette. 75 

Hi>lori,i Coehslls (1712). 41-2 

/ fotorfe CoektHs ( 1 725), 43. 44 

Hobbs, William. 53 

Hogarth. William. The Rake's Progress, 55. 56 

I lolkham Hall, 40 

Hooke. Robert. 25-7. 32 -3, 54. 58 

I lope-Jones, Frank, 169-70, 173, 233 

I lorrocks, Jeremiah. 21. 23 

Morton. Dr. 174 

Hoskins, Sir John and Lady, 33 

hours, equal and unequal, xv, 81 

I lubert, James, 53 

I luygens, Christiaan. 14. 15. 16, 33. 54 

Hyde Park, 28 

Hvdrographcrof theN'avy, 88, 163 

lllusimhtl London News, 87, 88. 95, 109. U3 

International Astronomical Union. 160. 166, 

International Telegraph Co.. 100 
invar. 218-19 
Irish Mail. the. 89 
fees. / '. frigate. 77 

Jacob staff. 6 

lames I, 23 

James II (Duke of York). 21. 41 

Janssen. Prof. J„ 139-40, 229 

Jenkins, David, 1 14 

Jerusalem, are meridians 

Jupiter's satellites, see Longitude 

Kant, Immanuel. 175 
Kelvin, see Thomson 
Kendall, Larcum, 71; Kl. 70, 71; K2. 711 
Kepler, Johann, 13 

Kerou.ille, Louise de. Duchess of Ports- 
mouth, 19, 2(1, 24-9, 69 
Keroualle, Sicur de, Set Penancoet 
Kikuchi, Prof., 229 
Knotty*, U'ttice. 31 

Lacaille, \icolas-Louis, Abbe de. 63 

I agrange, loseph-Louis, 78 

Lalande, J. -J. Lefrancais de. 73-5, 78 
I .mgley. Prof. S. P.. 121 
1 apetOUSe, Jean-Francois Calaupde. 77 
Laplace, Pierre-Simon. Marquis de, 78. 156, 

Latitude, 1,4, 192 
Leathern-apron Club, 59 

I cf.iivre. A„ 139-40. 229 

legal time, in France. 153-6. 186; Liberia, 156; 

Spain. 186: United Kingdom, 113-15; 

United States, 126 
I cicester, see Dudley 
l-emonnier. Pierre-Charles, 62 
Le Roy, Julien, 73 
l.e Roy, Pierre, 73, 75-8. 76 
Levasseur, M., 131 
Lewenhaupt, Count Girl, 230 
I ewisham Station. 90. 91, 100 
Lindley. Joseph, 116 
Liverpool Observatory, 116 
Lizard. The, MO meridians 
local mean time. 152 
Logan, James. 59 
London Bridge Station. 92, 94 

London District Telegraph Co.. 9-1, 101 
Longitude, xvi. 1, 193 

Act (1714). 42, 51-3,67-72 
Acts(1763. 1765, 1773), 69-72 
Board ol. 42. 52-3. 63-6. 78-9, 197 
Bureau des Longitudes. 78-9. 82. 156. 157 
counting of, 137, 144 

finding of. Appendix I; by 
sea, 9, II. 51. 53. 67, 198; by transport of 
chronometers, 1 16; by dead reckoning, 
16, 54; bv electric telegraph, 90-1. SB, 
116-18; by geodesv, 116. 170-1; bv 
Jupiter's satellites. 12. 16. 25. 51, 54, 60. 
64, 197-8; by lunar eclipse, 2, 6, 194. by 
magnetism. 25: by occupations. 4; by 
submarine cable. 117-18; Morin's 
method, 14, SI. Pierre's method. 25. 
Whiston and Ditlon's method. 48 9. 51. 


British, 51-2. 56. 67-72 

French, 16, 56 7; Rouille. 56-7. 58, 75-7 

I lolland. 12 

Portugal, 12 

Spain, 10 

Venice, 12 
scales on charts, 129, 130 
Longomontanus, 30 
Loran-C. 184 
I ouis IX (Saint Louis), 3 
Louis XIII. 128 
Louis XIV. 15. 16. 19. 25 
Louis XV. 75-7 
lunar distance. MY Longitude 
lunar tables, see Mayer 
lunar theorv, see moon's motion 
Lund. J. A.'. 104 
Lyons. Harold, 180 

Mackenzie, James Stewart. 73-4 
Madeira, sae meridians 
Magellan, Ferdinand, 160-1 
Mail io.ii Ill's, 83 

Marconi, Guglielmo, 163 

Marinus of Tyre. 127 

Marrison, Dr, 174 

Maskelvne, Kdmund, 64 

Maskclyne. Nevil, 63 7. 116, 197,212 

Massy, horologisl, 57 

Maudslav and Field, Messrs, 80 

Mayer. Tobias. 60, 62, 63-4, 71 

Mazarin, Hortense. Duchess of, 21 

Mazarin, Cardinal Jules, 14 

MeanSolarTime.38,67,82, 184, 187, 199 

Mean Sun, 107. 199 

Mech.iin. Pierre-Francois A.. 78 

Mercator, Gerhard. 128 

Meredith, Hugh. 59 

meridians, prime or initial, 127-31, 139-42: 
Amsterdam. 134; Azores, 128. 140; Benng 
Strait. 135. 140; Berlin, 140; Bethlehem. 
145; Brussels, 134; Cadiz. 141; Canary Is.. 

INDEX 251 

12m >>. Cape Verde Is.. 128; Christiania, 

134; Conferences. 128. 138 51; Copen- 
hagen. 134; Ferro, 128-9. 130. 134, 136: 
Fortunate Is., I, 127-8; Great Pyramid, 
Egvpt, 135, 140. 145; Greenwich, xiv. 45. 
66-7. 113. 129, 131, 134. 137, 139 42. 143. 
151, 170-1, 186-7, 189. 193; Jerusalem. 
Temple ol, 135, 140; Le Havre, 145; Lisbon. 
134; Lizard, the, 66. 129. I \0; London (St. 
Paul's). 129. 130; Madeira. 128; Munich. 
1.34; Pans. 129, 130. 131. 134. 136. 140. 143; 
Pulkowa. 134; Rhodes, 127; Rio de Janeiro. 
134; Rome. 134; Stockholm. 134; Tcncrifc, 

48, 128-9. (30. 140. loledo. 127; Warsaw, 

134; Washington, DC. 123, 140 
Mersains, astronomer, 77 
metric system. 136, 138.140 2 
mile, geographical, 51 n 
Mirelleur (Greenwich Castle), 31 
moon's motion, theory of. 61 
Moore. Sir |onas. 2 1 40 
Moore. W., 113 
Mi>n\ La, frigate, 77 
Morin.Jean-Baptiste. 14, 15.30 
Morland. Sir Samuel 23 
Morrell and Byers Is., 162 
Morion. Lord] 74-5 
Motel, Henri, 77 
Mudge, Thomas. 72, 75 
Mueller. Johannes, set Rcgiomontanus 

Nash, John, 96 

National Bureau of Standards (US), 180 

National I aboratorv (NPL). 173-4. 

180. 182, 221 
Nautical Almanac, 157-60; British. \iv, 60, 

64-7, 65, 83. 129 31. 149, 157. 163. 187. 

196-7; Russian, 131, US, 123; M cbe 

Comaissancetles Temps 

Navcsink. N'J. set time signals 
Navy Hoard. 32 

Neptune Francois, ants, 129, 130 

N'ewcomb. Simon, 135, 156. 157 
Newton, Isaac, 23, 41. 48, 50-1. 59. 60. 61 

New York Herald, 124 

N'evslett, -<v Reusner 
NitMM.USS, H7 
Niebuhr, Carslen. 63 
Nivcmois, Due de, 73 
N'orddcich. tee time signals 

Northcote. Sir Stafford. 109-12 

occultation. *v Longitude 

Octagon Room, s,v Greenwich 

odometer. 16 

Ordnance. Board (Office) of. 2 1 . 28, 32, 38. 40 

I, HMS, 68 
Orleans, Henrielte-Anne. Duchcssed'. 19 
Orleans. Philippe, Due d', 57 
Ortclius, Abraham. 128 
O'Shaughnessv. Dr, 92 
Osier. Abraham I-'.. 87, 105 




Oxford, Bad of, 49 

Oxford, lorn lower, 36. 109. HI 

I'lliln'm Ktlll, 128 

Palmer. John. 83 

Paris. Academie Rovale dos Sciences. 15, 16. 
18, 25, 56-7, 69. 73, 75 7. 129, Ecole 
Militaire. 78; Observatory. 15, 25, 78-9, 
90-1. m. 165-6, 170. 204; Paris Mean 

Time, 153; Treaty of. 69; teetho meridians. 

lime signals 
Parnell, Archdeacon Thomas. 49-50 
Parrv.Capl.J.F., RN, 165 
Parry. J. V. L, 180, 224 
Pasquier, E.. 153 
Paslorin. )uan, 230 
Paven, M., of Lorraine. 26 
Peil, Dr John, 25-9 
Penancoet. Guillaume, Sieur de Keroualle. 

pendulum: free-pendulum. 172; gridiron, 

172, 212. 2/3; invar. 218 19; mercurial. 172. 

209-12; simple, 13. 207. 2(17. vr tho clocks 
Pepvs, Samuel, 21 
Perfect, DrU. S.,204 
Pericles. 81 
Philip II, 127 
Philip III, 12 

Philippine Is. date kept. 160 I 
Photographic Zenith lube (PZT). 21)?. KB, 

Pic.ird. lean. 15. 16.25 
Pigafetta, Antonio, 160 
Pingre, Alexandre-Guv. 63, 77 
Playfair, Mr. 114 

Polar variation (wobble/wander), 177. 177-8 
Pole Star, 4. 192 

Pollock. Chief Baron ol the Exchequer, 1 14 
Pond, John, 84. 213; Me aba transit in- 
Pope, Alexander. 49-50 

Porfofaw, 3 

Portsmouth, Duchess of, Set Keroualle 
Pod Office, 83, 84. 86. 87, 96. 100-2: 
Research I aboratories, 174-5. 221 

/ w Oftuc C.ii;rf,\ 104 

Prasttn, Ducde, 73 

prime meridian, see meridians 

Ptolemy. Claudius, 1. 2. 127. 128. 192. 194 

Pulkowa Observatory, 1 16; ftralsa meridians 

Puttenham, George, 31 

quadrant. Hadlcy's reflecting, 58-60, 61. 62 
quadrature of the circle, 18. 54 
quartz, *v clocks 

Rabi. 1. 1. .224 

railroad, Mr railways 

railways and railroads: Caledonian, 88; 
Canadian National, 132; Canadian Pacific. 
132; Chester and Birkenhead, 88; Chester 
and Holyhead. 88; Fast Lancashire, 88; 

Edmburgh and Glasgow, 106; Great 
Western. 84, 87. 88, 97; Intercolonial 
(Canada), 132; Lake Shore and Michigan 
Southern. 120; Lancaster and Carlisle. 88; 
Liverpool and Manchester. 87; London 
and Greenwich, 84; London and North- 
western. 88; London and South-Westem. 
88; London. Chatham and Dover. 96; 
Midland, 87. 88: New York Central. 120; 
North Western, 87; Ontario. Simcoe and 
Huron, 132; Pennsylvania Central, 121; 
Pittsburgh, Fort Wayne and Chicago, 121; 
South Eastern, 87. 88. 89-92; York and 
North Midland. 88 

Railway Clearing House. 88 

railway time. 86-9. 105-13. HO. 187; Birm- 
ingham, 105; Bristol. 112-13; Detroit, 126; 
Fdinburgh, 106-7: Europe, 118 20; Fxeter, 
109-12; Glasgow. 106; Greenock. IIX); 
Liverpool, 106; Manchester, 106; Oxford. 
109; Perth, 106; Plymouth. 109, 113; 
Stirling, 106; USA and Canada, 120-6 

Ramsay, N„ 180 

Ramsden. Jesse, 60 

Raratonga, dale kept, 162 

Reflex Zenith lube, 204 

Regiomontanus (Johannes Mueller). 4. 5 

Keith. John. 169 

RewfllMOR, HMS. 71. 197 

Reusner, Andre, of Neystett. 16. 18, 54 

Rhodes, sir meridians 

Richelieu, Cardinal. 14. 128-9 

Richmond. 1st Duke ol. 19, 69; 3rd Duk. 01 

Riefier, Siegmund, 172-3. 215-16 

Robcrval. 16 

Rock. M. Miles. 229 

Rodgers. Rear-Admiral C. R. P.. USN, 138, 

Romer.Ole, 15,200 

Rummy, I IMS. 45 

Rouille. Prix, He I ongitude 

Roy, Gen. William, 116,171 

Royal Astronomical Society. 1^7 

Royal Commissions, 24 

Royal Greenwich Observatory. are 
I lerstmonceux 

Royal Navy, 72, 149-50 

Royal Observatory. tee Fdinburgh. Green- 
wich. Paris 

Royal Society. 15. 23. 41. 58. 166 

Rudd. K.J..173 

Rudolphine Cables. 13 

Rundle. R.. 113 

Rustem Effendi, 148. 230 

Rutherfurd. Lewis M.. 230 

Sabine. Col. Edward, 116 
St. Givryr. I IMS. 45 
St. Petersburg Academy. 62 
St. Pierre. Sieur de. 24-30 
Samoa, dale kept, 162 

Sampson. Cdr. W. T„ USN. 139-40. 230 
sand-glass. 10 
Saros s w le, M 

Sarum. Seth Ward, Bishop of. we Ward 

Scotsman. The, 106 

Scriblerus Club. 49 

second, atomic. 182, 224; ephemens, 179; 

leap. 183-7. 185; SI. 182. 224 
sextant, 57-60 
Shakerley. Jeremiah, 21 
Sharp, Abraham. 44 
Shelton, John. 69, 2 JO 
Shepherd, Prof. Anton, 61 
Shepherd, Charles, 90-1. 93, 97. 213-15. 214 

Sherwood, J. E.. 180 

Short. James. 75 

Shortt. W. II.. 172-3. 210, 216-19. 217. 218 
Shovel. Sir Clowdislcv. 45-7. 46 
sidereal time, 38. 200 

Smyth, Prof. Pia//i. 135 

Soteldo, Snr Dr. A. M ,230 

Southwell. Edward. 50 

Speaking Clock, set time signals 

Spence. Joseph. 50 

Spencer Jones, Sir Harold. 175. 178 

Si./i»n,HMS. 118 

standard time. 121-6. 125, 135, 137. 152-6. 

158-9; at sea. 169 
Standard Time Co.. 104 
Stanhope, Charles. 68 
Stanhope, James. 51. 68 
Statutes (Definition of Time) Act (1880). I II 
Stebnitzki. Maj.-Gen.. 230 
Steele, Sir Richard, 48 
Stewart, CapL John, 230 

Stdffler. John. 6 

Strachev. Lt.-Cen. R.. 139, 142, 229 

Strati, Rt. Hon. Fdw.ud.87 

Strove, Prof. O.. 156 

submarine cable, 117; Atlantic, 117-18. N"; 

Channel, 90. 1 17: mv nisei ongitude 
Summer Time. StV Davlight Saving Time 
sundial. 81-2, 187 
Swift, Jonathan, 49 90, =*, 
SynchronorneCo. Ltd., 173 

TAI, tee Atomic rime. International 

TurMr. HMS, 69 

Teddington. *v National Physical I aboralory 

Telegraph Construction and Maintenance 

Co.. 117 
Tenerife. 48; SeeabO meridians 

lerrestrial nine. 132, 152 
Terribk, HMS, 118 

Themislocles. 78 

Thomson. Prof. William, lord Kelvin. 118. 

Tietjen, Prof.. 156 
Tilbury Fort. 32 
time. Me inilii-iilual lime scabs ET. GMT. TAI. 


time-balls: Carter. Old Kent Road London, 

103, KM; Comhill. I ondon. 95. 96. 100: 
Deal, 94, 99. IIX); Fdinburgh, HXI; Green- 
wich, 76. 79-80. 100. 188, 213; I iverpool. 
100; New York, 125; Si l Mens, B0; small, 
102. 103, 104. Sirand. London, 92, 96 7. 
time distribution service. 
in France, 105 

in Great Bntain. 89-94, 99-102, 173. 180. 
188. 221; by hand. 83 6; for private 

subscribers, 104-5 

in United States, 121, 125. 180 
lime-finding atnient, SI, by astronomy. 

Appendix II, by double altitudes. 200; bv 

Photographic Zenith lube. 202, 203, 204; 

in prismatic astrolabe, 2"', 204; bv transit 

telescope. 143. 200-2 
time-guns: Dover, 99; Edinburgh. 11X1; 

Newcastle and South Shields. 101 

tim ek eeper s , marine, we Chronome t er s 

time scales: atomic. 182; RGO (GA), is I 2; 
USNO(Al), 182 

time signals: hv radio, 163-5, 168: bv tele- 
graph, 89-105. 121. 164; bv telephone 
(Speaking Clock), 170; early US, 163, Intel 
Tower, Pans, 164 5; MSP, WWV. 180. 184: 
Norddeich, 163-5; private subscribers. 
UK, 102. 103. 104-5. rhvthmic (vernier), 
I64i Rugby, 115, 164; six pip. 169. 175 

lime Zone Chart, 158-9 

Times. Die. 92, 105-6,107-9. 114 

Titus. Col. Silius. 25-7 

Toledo. Mr meridians 

Tompion. Thomas, 33, 209 

Tordesillas. Treaty of (1494), 127 

Tower of London', 21 , 30, 32, 40 

FowneJey, Chri st opher, 21, 23 

Townelev. Richard. 21. 13. 36-8 

transit circle, 202; Airv's. xiv, 84, 139, (43. 

170-1, 189. 202 
transit instrument, 200. 201; Bradley's. 171; 

Ponds. 171 
I'riiiiniii. sloop. 59 
Tupper. Alvaro Bianchi. 230 
I wain. Mark. 6 

US Naval Observatory, tee Washington 
Universal Character, lip. Wllkins's, in 
Universal lmie(UT),xlv xv,137, 144 8, isj. 

160, 178. 187-90; Co-ordinated Universal 
Time (UTC), 183-6, 185, 190; LTD. UTt, 
UT2. 178-9, 182. 183-4. 185. 201, fee ttko 
day. universal 

Valentia Island, lr.-l.ind, 117. 118. 119 
Valera. Juan, 230 
Yarley.C P., 94, 99, 101 

Venus, transit of. 63 
Vespucci. Amerigo. 6 
Victoria. Queen. 90 

..;. ship. 160 
Von Alvensleben. Baron 1 1., 229 


i;ki iawuii n\ii 

Von Moltke. Count, 120 
VonSchaeffer, Baron I.. 229 
Vullfamy. B. I . iin 

Wile-. William. 71. 149. 197 

Walker, C V., 89-92, 99 

Walker, 5. C, 117 

Walther, Bernhard, 13 

Want, Sefh, Bishop ol Saturn, 24-7 

Washington, DC Naval Observatory, 121 

180. 2M;scva/si> meridians 
Watson, Baron of the Exc h equer, 113-44 
Wauchope, Opt Robert, 79 
Werner, Johann, 6, 57, 194 5 
Western luminary, 112 

Wharton Sir George. 32 

Wharton, Mycarpua, .12 
Wheatstone, Charles, B3 

Wheeler. Adm., 47 

Whteton, William. 46. 47-50. 56 

Wilkins. Bishop. It 

Uilk-ii. WUHam, 166-7 

William 111, 41 
Williams, Francis. 53 
Williamson. Sir Joseph, 25, 29 
n„:;,)ston.DrW. H..87 
Wren, Sir Christopher. 16. 23. 28. 31. 32. 36, 

vear: leap year, 184, 187; length of. changes 

in, 178 
York. Duke of, see James II 

Za d ha ri as , ). R.. 180 

/one lime, Stl Standard Time 


Man and the Stars 

by Hanbury Brown 

The Oxford Companion to Ships and 
the Sea 

edited by Peter Kemp 

jacket photo: Marie Kenny, courtesy of the Trustees, 
National Maritime Museum, Greenwich. 


ISBN 019 215948 8